Molecular Cancer Research Infection and Cancer: Biology, Therapeutics, and Prevention 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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Breuleux, M.
Right arrow Articles by Eppenberger, U.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Breuleux, M.
Right arrow Articles by Eppenberger, U.
Molecular Cancer Research 4:27-37 (2006)
© 2006 American Association for Cancer Research

Signaling and Regulation

Heregulins Implicated in Cellular Functions Other Than Receptor Activation

Madlaina Breuleux, Fabrice Schoumacher, Daniel Rehn, Willy Küng, Heinz Mueller and Urs Eppenberger

Molecular Tumorbiology Unit, Department of Research, Stiftung Tumorbank and University Clinics Medical School, Basel, Switzerland

Requests for reprints: Madlaina Breuleux, Novartis Pharma AG, Klybeckstrasse 125, WKL-125.12.59, 4002 Basel, Switzerland. Phone: 41-61-696-25-17; Fax: 41-61-696-63-81. E-mail: madlaina.breuleux{at}novartis.com


   Abstract
Top
 Notes
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Heregulins (HRG) are known as soluble secreted growth factors that, on binding and activating ErbB3 and ErbB4 cell surface receptors, are involved in cell proliferation, metastasis, survival, and differentiation in normal and malignant tissues. Previous studies have shown that some HRG1 splice variants are translocated to the nucleus. By investigating the subcellular localization of HRG{alpha}1-241, nuclear translocation and accumulation in nuclear dot-like structures was shown in breast cancer cells. This subcellular distribution pattern depends on the presence of at least one of two nuclear localization sequences and on two domains on the HRG construct that were found to be necessary for nuclear dot formation. Focusing on the nuclear function of HRG, a mammary gland cDNA library was screened with the mature form of HRG{alpha} in a yeast two-hybrid system, and coimmunoprecipitation of endogenous HRG was done. The data reveal positive interactions of HRG{alpha}1-241 with nuclear factors implicated in different biological functions, including transcriptional control as exemplified by interaction with the transcriptional repressor histone deacetylase 2. In addition, HRG{alpha}1-241 showed transcriptional repression activity in a reporter gene assay. Furthermore, a potential of HRG proteins to form homodimers was reported and the HRG sequence responsible for dimerization was identified. These observations strongly support the notion that HRG1 splice variants have multifunctional properties, including previously unknown regulatory functions within the nucleus that are different from the activation of ErbB receptor signaling. (Mol Cancer Res 2006;4(1):27–37)


   Introduction
Top
 Notes
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
The heregulin (HRG) family of growth factors plays an important role in regulating cell proliferation, metastasis, differentiation, and survival of various normal and neoplastic tissues (1). All human HRGs originate from four genes, HRG1, HRG2, HRG3, and HRG4, respectively (2-4). Alternative RNA splicing results in 15 HRG isoforms, which vary in their mosaic-like composition of different functional domains. Most HRGs are soluble, secreted 44-kDa glycoproteins originating from transmembrane precursors, undergoing typical glycosylation and trafficking (5). The secreted extracellular region of HRGs contains a C2-type immunoglobulin (Ig)–like domain (exons 3 and 4), which binds to extracellular matrix proteins containing glycosaminoglycan chains (6), and an epidermal growth factor (EGF)–like domain required for ErbB receptor binding and activation (1, 7). However, the very NH2-terminal region (exons 2 and 3) exhibits also a nuclear localization signal (8).

Secreted HRG proteins act as ligands for some members of the ErbB family of class I receptor tyrosine kinases consisting of EGF receptor, ErbB2 (HER-2/neu), ErbB3, and ErbB4. The expression levels of ErbB1 to ErbB4 have an effect on HRG response; furthermore, the precise signaling cascades and biological responses evoked by HRG proteins and their receptors are cell type specific (9). This finding is important to define the tumorigenic role of ErbB receptors and HRGs in various cancers. Analysis of cellular expression patterns of HRGs and their transmembrane receptors indicate that the signaling can be paracrine or autocrine in nature (10).

Up-regulation of HRG was shown to be sufficient for the development of mammary tumors independently of estrogen stimulation and ErbB2 overexpression (11). Moreover, inhibition of HRG expression suppressed the aggressive phenotype by decreasing ErbB activation and reducing matrix metalloproteinase-9 activity (12). These data show a direct causal role of HRG in the induction of tumorigenicity. Therefore, HRG could be a key promoter of breast cancer tumorigenicity and metastasis independently of ErbB2 overexpression.

HRGs do not only act by initiating surface receptor-mediated signaling but may also be involved in alternative signaling pathways. In this respect, receptor-bound HRG1ß1 was shown to be transported to the nucleus (13), independent of nuclear receptor translocation. Nuclear HRG1ß1 modulated the activity of c-myc, a critical regulator of cell cycle progression, differentiation, and malignant transformation (13), as well as stimulated cancer cell proliferation in vitro (8). Another HRG splice variant, HRG1ß3, lacks the transmembrane domain and the cytoplasmic tail and is not secreted; however, the nuclear localization sequence (NLS) in its NH2 terminus suggests that it accumulates in the nuclear compartment (8). Importantly, nuclear localization of the HRG precursor was shown in papillary thyroid carcinomas but not in normal thyroid tissue (14) and was not associated with the expression of ErbB receptors. Nuclear staining for HRG has also been observed in medulloblastomas (15). Furthermore, HRG1{alpha}, HRG1ß, and HRG3 were recently shown to localize to the nucleus in ductal carcinoma in situ of the breast (16). Several other growth factors act not only as extracellular ligands for transmembrane receptors but also as nuclear growth factors. Schwannoma-derived growth factor and fibroblast growth factor-1 were reported to depend on a NLS to achieve mitogenic activity (17, 18). Heparin-binding EGF-like growth factor, a member of the ErbB receptor ligand family, is also translocated to the nucleus and plays a role in disease progression in various cancers (19).

Due to the nuclear localization of HRG proteins and their molecular diversity, questions have arisen concerning the functional multiplicity of HRG isoforms in the various tissues. Recently, it has been shown that NRG1ß3 localizes to two known intranuclear structures, nucleoli and SC35-positive nuclear speckles (20), independent of the receptor-binding domain or the previously predicted NLS.

In our study, we have characterized the nuclear localization of HRG{alpha}1-241, a HRG1 isoform, which shows sequence homology with NRG1ß3, in breast cancer cells. Assessing the structural requirements of HRG{alpha}1-241 for nuclear translocation and specific subnuclear localization with a more detailed analysis using different HRG{alpha}1-241 deletion variants, a second putative NLS within the Ig-like domain of HRG{alpha}1-241 was revealed. Moreover, two sequences responsible for subnuclear dot formation were identified. Focusing on the nuclear function of HRG, a mammary gland cDNA library was screened in a yeast two-hybrid assay to isolate potential nuclear protein interaction partners of HRG{alpha}1-241. Positive interactions of HRG with several nuclear proteins were shown, including the interaction with the transcriptional repressor histone deacetylase 2 (HDAC2), which was confirmed by coimmunoprecipitation of endogenous HRG in breast cancer cells. Furthermore, a transcriptional repression activity of HRG{alpha}1-241 was revealed using a reporter gene assay. In addition, a potential of HRG proteins to form homodimers was reported and the HRG sequence responsible for dimerization was identified. These results strongly support the hypothesis that HRG proteins modulate cellular functions by nuclear mechanisms independent of receptor activation.


   Results
Top
 Notes
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 Acknowledgements
 References
 
Nuclear Expression of HRG in Breast Cancer Cells
To determine whether endogenous HRG was expressed in the nucleus of breast cancer cell lines, nuclear and cytoplasmic protein fractions were prepared and analyzed by Western blot (Fig. 1 ). All cell lines tested expressed full-length HRG{alpha} protein (44 kDa) in the nucleus as well as in the cytoplasm; however, the relative expression levels differed for each cell line (Fig. 1, top). In addition, MDA-MB-231 cells showed nuclear expression of shorter HRG forms (~30 kDa), probably either nonsecreted isoforms lacking part of the EGF-like domain (e.g., HRG1ß3 and HRG1{gamma}) or the secreted extracellular domain of HRG (Fig. 1, top middle). The purity of nuclear and cytoplasmic protein fractions was confirmed using antibodies against the nuclear protein poly(ADP-ribose) polymerase and the cytoplasmic protein MEK-1, respectively (Fig. 1, bottom). These data clearly confirm the nuclear localization of endogenous HRG in breast cancer cell lines, consistent with previous reports on breast cancer biopsies (16).


Figure 1
View larger version (36K):
[in this window]
[in a new window]
  		FIGURE 1. Endogenous HRG expression in breast cancer cell lines. Nuclear (N) and cytoplasmic (C) protein fractions were subjected to immunoblotting using a polyclonal anti-HRG antibody. Poly(ADP-ribose) polymerase (PARP) and MEK-1 detection are used as a control for the purity of the nuclear and the cytoplasmic fraction, respectively. Full-length HRG (44 kDa) as well as either the secreted extracellular domain of HRG or shorter intracellular HRG isoforms (~30 kDa) can be found in the nuclear fraction of the different tumor cell lines at different expression levels.

 
Nuclear Import and Nuclear Distribution Are Dependent on Different HRG Domains
To show the nuclear translocation of HRG{alpha}1-241, MCF-7 breast cancer cells were transfected with pEGFP/HRG{alpha}1-241 and analyzed 24 hours after transfection. Transfection of the control plasmid pEGFP-C1 resulted in uniform distribution of the green fluorescent protein (GFP) throughout the entire cell (Fig. 2A ). Nuclear accumulation of GFP was not observed, indicating that GFP alone is not imported into the nucleus. Fusing the sequence of a minimal nuclear localization signal of the SV40 large T antigen to pEGFP-C1 led to the efficient nuclear import of the fusion protein with diffuse nuclear staining (Fig. 2B). Strikingly, the GFP/HRG{alpha}1-241 accumulated predominantly in the nucleus and formed discrete nuclear dots (Fig. 2C).


Figure 2
View larger version (72K):
[in this window]
[in a new window]
  		FIGURE 2. Subcellular localization of NH2- and COOH-terminally deleted HRG{alpha}1-241 fusion proteins after transfection into MCF-7 cells. A. Cellular distribution of GFP alone as control for cytoplasmic distribution. B. Cellular and nuclear localization of GFP fused to the NLS of the SV40 large T antigen as a control for nuclear accumulation. C. Nuclear dot-like structure formation by wild-type HRG{alpha}1-241 as well as by HRG{alpha}1-148{Delta}BbsI, HRG{alpha}1-187{Delta}XmnI, and HRG{alpha}1-205{Delta}XhoI (data not shown). D. Nuclear retention of HRG{alpha}1-26{Delta}SacII as well as HRG{alpha}1-52{Delta}SpeI and HRG{alpha}1-113{Delta}BclI (data not shown). E. Accumulation of {Delta}SacII-HRG{alpha}28-241 in nuclear dots. F. Accumulation of {Delta}SpeI-HRG{alpha}53-241 in the nucleus. G. Cytoplasmic distribution of {Delta}BclI-HRG{alpha}114-241 as well as {Delta}BbsI-HRG{alpha}150-241 and {Delta}XmnI-HRG{alpha}188-241 (data not shown). Bar, 10 µm.

 
To determine which HRG{alpha}1-241 domains are responsible for the nuclear import and subnuclear localization into dot-like structures, the HRG{alpha} sequence in the pEGFP/HRG{alpha}1-241 was progressively deleted from both the 5' and the 3' ends (Fig. 3 ). Fusion proteins with deletions in the HRG{alpha}1-241 COOH terminus revealed two different localization patterns. The fusion proteins of pEGFP/HRG{alpha}1-205{Delta}XhoI, pEGFP/HRG{alpha}1-187{Delta}XmnI, and pEGFP/HRG{alpha}1-147{Delta}BbsI localized in dot-like structures in the nucleus, comparable with the expression of wild-type GFP/HRG{alpha}1-241 (Fig. 2C). In contrast, GFP/HRG{alpha}1-26{Delta}SacII fusion proteins were imported into the nucleus, but without the formation of distinct nuclear dots (Fig. 2D), comparable with the localization pattern of the NLS-coupled GFP (Fig. 2B). HRG{alpha}1-52{Delta}SpeI and HRG{alpha}1-113{Delta}BclI showed the same localization pattern as GFP/HRG{alpha}1-26{Delta}SacII (data not shown). Transfection of pEGFP/HRG{alpha} constructs with progressive deletions in the NH2 terminus of HRG{alpha}1-241 revealed three different cellular localization patterns. Transfection with pEGFP/{Delta}SacIIHRG{alpha}28-241 resulted in a wild-type localization pattern (Fig. 2E), whereas cells transfected with pEGFP/{Delta}SpeIHRG{alpha}53-241 showed a nuclear accumulation of this fusion protein without dot formation (Fig. 2F). GFP/{Delta}BclIHRG{alpha}114-241, {Delta}BbsIHRG{alpha}150-241, and {Delta}XmnIHRG{alpha}180-241 fusion proteins were distributed throughout the cell similar to the distribution pattern of GFP alone (Fig. 2G). These results clearly show that the domains responsible for nuclear import are distinct from those involved in dot formation, with both being localized at the NH2 terminus of HRG{alpha}1-241.


Figure 3
View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 3. Nuclear import and subnuclear localization of HRG constructs. All HRG{alpha}1-241 fusion proteins were assessed for their capacity to be imported into the nucleus and/or to be localized in nuclear dot-like structures.

 
Nuclear Import and Localization Do Not Depend on the EGF-Like Domain of HRG{alpha}
Categorization of the GFP/HRG{alpha} fusion proteins according to their capacity to be imported into the nucleus and/or to be confined to dot-like structures is shown in Fig. 3. Two putative nuclear localization signals that are required for nuclear import or nuclear retention of the fusion proteins were found in HRG{alpha}1-241 (Fig. 4A ). The NH2-terminal NLS is located at amino acid positions 12 to 18 and was described by others as a putative NLS for HRG proteins (8). The second putative NLS is located within the Ig-like domain, showing no homology to one of the three types of NLS described earlier (21). Either of the two NLS is sufficient for nuclear import. In contrast, for nuclear dot formation, both of the two dot-forming sequences found in HRG{alpha}1-241 are required (Fig. 4A). The first of these domains is encoded in exon 2/3 (amino acids 28-52) corresponding to a stretch of 25 amino acids NH2-terminal of the Ig-like domain, and the second is in exon 4/5 (amino acids 114-147), which corresponds to the 34–amino acid spacer region located between the Ig-like domain and the EGF-like domain. Deletion of one of these two domains resulted in the loss of subnuclear dot formation (Fig. 3).


Figure 4
View larger version (59K):
[in this window]
[in a new window]
  		FIGURE 4. Nuclear dot formation by HRG{alpha}1-241 does not require the EGF-like domain. A. Two putative NLS (gray boxes) and two regions (black boxes) required for the localization in nuclear dots. B. Minimal HRG{alpha}1-241 sequence required for nuclear import and dot formation. C. MCF-7 cells transfected with the minimal HRG{alpha}1-241 sequence fused to GFP. Hatched boxes, GFP tags; light gray boxes, HRG{alpha}1-241 sequences; dark gray circles, cysteine positions. Bar, 10 µm.

 
Thus, the minimal requirements for HRG{alpha}1-241 to be imported into the nucleus and to form nuclear dot-like structures are the presence of either one of the two NLS and both dot-forming sequences located on both sides of the Ig-like domain. To confirm this notion, we constructed a minimal HRG fusion protein (pEGFP/{Delta}SacIIHRG{alpha}28-147{Delta}BbsI) consisting of the two domains responsible for nuclear dot formation on either side of the Ig-like domain and the Ig-like domain containing just one of the two NLS (Fig. 4B). As predicted, the intracellular distribution of this fusion protein was comparable with that of wild-type HRG{alpha}1-241 (Fig. 4C). This finding clearly indicates that the EGF-like domain is not involved in directing HRG{alpha}1-241 into the nucleus.

HRG{alpha}1-241 Interacts with Known Nuclear Proteins
To further evaluate the potential role of intracellular and nuclear HRG{alpha}1-241, a yeast two-hybrid screen was done to detect novel protein interaction partners. A human mammary gland cDNA library fused to the GAL4 activation domain, containing 3.5 x 105 independent clones, was screened with HRG{alpha}1-241 fused to the GAL4-DNA-binding domain (DBD; ref. 22). Three different reporter genes (ADE2, HIS3, and lacZ) were used to detect protein interactions in the GAL4-responsive yeast strain AH109, each under the control of distinct GAL4 upstream activating sequences and TATA boxes. HRG{alpha}1-241 was unable to activate reporter gene expression by itself (data not shown). Of a total of 1,013 positive clones, which were picked and back-transformed either with or without HRG{alpha}1-241 to confirm the need for HRG{alpha}1-241 interaction to activate the reporter genes, 360 clones remained positive for their interaction with HRG{alpha}1-241 and were used for further sequencing analysis. Thirty-one different proteins implicated in multiple cellular functions were identified and further tested for domain specificity of their interaction with HRG{alpha}1-241 (data not shown). To identify HRG domains needed for interaction, deletion variants of HRG{alpha}1-241 were constructed either coding for the Ig-like domain (amino acids 1-128), the EGF-like domain (amino acids 112-241), or the minimal sequence shown to be sufficient for nuclear import and subnuclear dot formation (amino acids 21-150; Fig. 5A ).


Figure 5
View larger version (34K):
[in this window]
[in a new window]
  		FIGURE 5. Relative strength of protein-protein interactions with HRG{alpha}1-241. A. Schematic representation of HRG constructs. Gray boxes, Ig-like domain and EGF-like domain; black boxes, two putative NLS; hatched boxes, two domains necessary for nuclear dot formation; black bars, HRG{alpha}1-241 subfragments, each showing its respective position in the protein and its length. DFS, dot-forming sequences. B to D. HRG{alpha}1-241 (black column) and its deletion variants (dotted column, Ig-like domain; hatched column, EGF-like domain; squared column, minimal sequence) were cotransformed with the respective clones and the ß-galactosidase units were expressed as relative values compared with an internal positive control for protein interaction. As a background control, the empty yeast vector pGBKT7 was cotransformed in AH109 yeast together with the different clones to look for HRG{alpha}1-241 independent activation of the ß-galactosidase reporter gene (white column). All experiments were done three times in duplicates. B. Cullin-1. C. MDGI. D. G/T mismatch-specific TDG. E. hUBC9.

 
A summary of the key interacting proteins is presented in Table 1 . The interaction of HRG{alpha}1-241 with the nuclear proteins RS cyclophilin, serine/arginine nuclear matrix protein, and RNA helicase, proteins required for pre-mRNA splicing, is supported by previous data showing that NRG1ß3 was found to associate with specific intranuclear compartments implicated in RNA splicing (20). Therefore, these findings serve as a control for the robustness of the data obtained by the yeast two-hybrid screen.


View this table:
[in this window]
[in a new window]
  		Table 1. Result of Yeast Two-Hybrid Screen

 
Having done this detailed analysis of the positively interacting clones, a subset of these candidates was chosen for further studies, focusing on potential nuclear interaction partners of HRG. HRG{alpha}1-241 was found to interact strongly and in a domain-specific manner with Cullin-1 (Genbank AF62536), mammary-derived growth inhibitor (MDGI; Genbank Y10255), G/T mismatch-specific thymine DNA glycosylase (TDG; Genbank HS51166), and human ubiquitin-conjugating enzyme 9 (hUBC9; Genbank HSUBC9ENZ), which are involved in diverse biological processes. A liquid ß-galactosidase assay was applied to quantitate the relative strength of positive protein interactions with HRG{alpha}1-241 and its deletion variants (Fig. 5B-E). To compare the results of different assays, all values were normalized to a mean number of ß-galactosidase activity obtained in a simultaneously performed internal positive control for protein interaction.

Cullin-1 revealed a clear interaction specificity for the EGF-like domain of HRG{alpha}1-241 (Fig. 5B). Although MDGI was able to slightly activate the ß-galactosidase reporter gene without the need of a positive interaction with HRG{alpha}1-241, the interaction with HRG{alpha}1-241 and with the EGF-like domain showed enhanced activity (Fig. 5C), indicating specific protein interaction with parts of HRG{alpha}1-241. G/T mismatch-specific TDG interacted specifically with the Ig-like domain of HRG{alpha}1-241, whereas the interaction with the EGF-like domain was comparable with the background activation of the reporter gene (Fig. 5D). Although hUBC9 was not able to activate the reporter gene without HRG{alpha}1-241 interaction, no clear domain specificity for either one of the deletion variants of HRG was detected; hUBC9 seems to interact with both the Ig-like domain and the EGF-like domain. The interaction was much stronger if both of the dot-forming domains were present in the construct (Fig. 5E).

Coimmunoprecipitation experiments with the candidate proteins and HRG{alpha}1-241 were done to confirm the data obtained with the yeast two-hybrid analysis. Cos-7 cells were transiently cotransfected with pcDNA3.1 plasmids, containing either the respective full-length cDNA of the clones identified in the two-hybrid screen fused to a hemagglutinin (HA) tag or the HRG{alpha}1-241 fused to a myc tag. As a negative control, pcDNA3.1/myc was cotransfected with pcDNA3.1/HA (data not shown). Cullin-1 (Fig. 6A ), MDGI (Fig. 6B), and G/T mismatch-specific TDG (Fig. 6C) coprecipitated together with HRG{alpha}1-241. Only hUBC9 did not show any interaction with HRG{alpha}1-241 (Fig. 6D) with this experimental approach. Hence, most of these results confirm the interaction of HRG{alpha}1-241 with nuclear proteins found in the yeast two-hybrid screen.


Figure 6
View larger version (51K):
[in this window]
[in a new window]
  		FIGURE 6. Association of proteins with HRG{alpha}1-241. Cos-7 cells were either cotransfected with myc-tagged HRG{alpha}1-241 and the respective HA-tagged Cullin-1 (A), MDGI (B), G/T mismatch-specific TDG (C), and hUBC9 (D) or cotransfected with different myc-tagged and HA-tagged HRG domains (E-H). To look at HRG dimerization, cotransfections were done with the following combinations: (E) myc-tagged HRG{alpha}1-241, HA-tagged HRG{alpha}1-241, (F) myc-tagged HRG{alpha}1-241, HA tagged Ig-like domain of HRG{alpha}1-128, (G) myc-tagged minimal sequence of HRG{alpha} (amino acids 21-150), HA-tagged HRG{alpha}1-241. H. Myc-tagged HRG{alpha}1-241 was cotransfected with HA-tagged EGF-like domain (amino acids 121-241; lanes 1 and 2) and with HA-tagged EGF-like domain coupled to a NLS (lanes 3 and 4). Blots are overexposed to show that no interaction takes place between HRG{alpha}1-241 and the EGF-like domain. The lysates were precipitated with a monoclonal anti-myc antibody. The supernatants (Sup) and the immunoprecipitates (IP) were subjected to immunoblotting using a polyclonal anti-HA antibody. Right, molecular size markers. Asterisks, expected sizes of the proteins.

 
HRG Forms Homodimers via Its NH2 Terminus
Homodimerization of proteins could be one mechanism inducing nuclear dot formation. Furthermore, the activity of HRG might be regulated via homodimerization. Therefore, we used the yeast two-hybrid system to test HRG dimerization. The respective coding sequences of HRG{alpha}1-241 and its deletion variants were subcloned in-frame into yeast two-hybrid vectors containing the GAL4-DBD (bait vectors) and the GAL4 activation domain (prey vectors), respectively. For direct testing of the interaction between HRG fusion proteins, each bait plasmid was paired with each prey plasmid and the interaction was assessed qualitatively (ADE2 and HIS3) and quantitatively (lacZ) by determining reporter gene activity in cotransfected AH109 cells. All HRG{alpha}1-241 fusion proteins containing the Ig-like domain were found to form homodimers, whereas interactions with the EGF-like domain of HRG{alpha}1-241 did not take place (data not shown).

Coimmunoprecipitation was done confirming HRG{alpha}1-241 homodimerization via its NH2 terminus (Fig. 6E-G) but not via the EGF-like domain (Fig. 6H, lanes 1 and 2). However, the construct containing the EGF-like domain is homologous to the HRG protein encoded by pEGFP/{Delta}BclIHRG{alpha}114-241, which does not contain a NLS and was shown not to be translocated to the nucleus. Cloning a NLS from the SV40 large T antigen to the HA-tagged EGF-like domain (Fig. 6H, lanes 3 and 4) did not lead to coimmunoprecipitation with HRG{alpha}1-241, further confirming that the dimerization of HRG{alpha}1-241 is independent of the EGF-like domain.

HRG Interacts with HDAC2 and Shows Potential Transcriptional Repression Activity
Many proteins implicated in transcriptional control are able to form homodimers. The homodimerization of HRG and its specific interaction with nuclear proteins suggest, therefore, a role for HRG in transcriptional control. To assess the interaction of endogenous HRG with HDAC2, a protein shown to interact with HRG{alpha}1-241 in the yeast two-hybrid screen (Table 1) and which is a well-known enzymatic transcriptional corepressor (23), we did a coimmunoprecipitation analysis in MDA-MB-231 cells and were able to coprecipitate HDAC2 with HRG. These results show clear evidence for an interaction of endogenous HRG with one of the proteins found in the yeast two-hybrid screen in a natural setting (Fig. 7A ); furthermore, these data support a transcriptional regulation function of HRG.


Figure 7
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 7. HRG{alpha}1-241 interacts with HDAC2 and shows transcriptional repression in a luciferase reporter gene assay. A. In vivo interaction of HRG{alpha} with HDAC2 shown by coimmunoprecipitation. B and C. Luciferase reporter gene was either under the control of a constitutively active GAL4 promoter in the pGAL4(5xUAS)-SV40-Luc plasmid (B) or under the control of a minimal GAL4 promoter in the pGK-1 plasmid (C). Luciferase activity was measured either alone (black columns, background activation), in combination with overexpressed HDAC2 (dotted columns), or in the presence of HRG{alpha}1-241 coupled to the GAL4-DBD without (squared columns) or with (hatched columns) HDAC2. When no HRG{alpha}1-241 was transfected, the plasmid containing the GAL4-DBD was cotransfected with the respective luciferase reporter gene plasmids with or without HDAC2.

 
Hence, a transcriptional study was done using a plasmid containing the luciferase reporter gene either under the control of a constitutively active GAL4 promoter [pGAL4(5x UAS)-SV40-Luc] or with a minimal GAL4 promoter (pGK-1). We transfected the DBD of GAL4 or HRG{alpha}1-241 coupled to the DBD of GAL4 together with either one of the luciferase plasmids and could show that HRG{alpha}1-241 was able to repress luciferase expression independently of the promoter status (Fig. 7B and C). Moreover, we transfected Cos-7 cells with a plasmid expressing HDAC2. HDAC2 was transfected either alone or in combination with HRG{alpha}1-241; however, it did not potentiate the transcriptional repression seen with HRG{alpha}1-241, which may already show maximal repression potential in this reporter gene assay. Our results clearly suggest that HRG has transcriptional control activity.


   Discussion
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 Acknowledgements
 References
 
Mitogenic growth factors are generally cell surface–associated or secreted proteins, which produce effects by binding to cell surface receptor tyrosine kinases. Evidence is mounting, however, that growth factors and members of the type I family of transmembrane receptors have also an important role directly within the nucleus. At present, there are few hypotheses suggesting nuclear functions of growth factors or their receptors (24).

Nuclear Translocation of HRG{alpha}1-241
NRG1ß3 was shown recently to localize to the nuclei of human breast cancer cells in a receptor-independent way, supporting the idea that secretion and subsequent cell surface receptor binding of HRG proteins are not a prerequisite for nuclear localization and that nonsecreted ligands may have highly specific functions in defined nuclear compartments, such as the nucleoli and SC35-positive nuclear speckles (20). Further evidence for a nuclear function of HRG proteins during malignancy arises from studies showing that HRG is expressed in the nucleus of breast as well as thyroid cancer biopsies and medulloblastomas (14-16), whereas no nuclear HRG was found in normal tissues. We have shown that full-length HRG and either extracellular HRG or shorter intracellular isoforms of HRG are present in the nucleus of different breast cancer cell lines. Furthermore, we have shown clear evidence for HRG{alpha}1-241 being translocated to the nucleus, where it is localized in dot-like structures. Several of our HRG{alpha}1-241 deletion constructs (such as HRG{alpha}1-205{Delta}XhoI, HRG{alpha}1-187{Delta}XmnI, and HRG{alpha}1-148{Delta}BbsI) lacked a portion of the EGF-like domain, shown to be necessary and sufficient for receptor binding and activation (7), yet they were imported into the nucleus and able to form nuclear dot-like structures. Thus, it is conceivable that the nuclear translocation of our constructs did not depend on ErbB3 and ErbB4 binding and activation, confirming the data of Golding et al. (20), showing nuclear localization of NRG1ß3 in a receptor-independent way. Our demonstration that HRG{alpha}1-241 is translocated to the nucleus suggests that novel mechanisms of action of HRG different from receptor binding and activation might exist, whereby HRG may also act as an intracrine growth factor, potentially extending the biological functions of HRG proteins.

HRG{alpha}1-241 Contains Two NLS
Certain proteins are transported actively and selectively into the nucleus if they contain a nuclear localization signal or are associated with NLS-containing proteins. In HRG{alpha}1-241, two short peptides were defined as active nuclear targeting sequences. The first NLS (KGKKKER) is located at the NH2 terminus of the mature (secreted) HRG, whereas the second NLS is located within a region of 16 amino acids in the Ig-like domain of HRG and resembles the consensus NLS sequence K-R/K-X-R/K (25). Our HRG constructs transfected into breast cancer cells gave rise to proteins smaller than the diffusion size limit for the nuclear envelope (~40-60 kDa). Therefore, nuclear import of HRG may occur by passive diffusion with nuclear retention or binding to nuclear proteins in a subnuclear compartment causing the dot-like structures. However, because small HRG{alpha}1-241 constructs lacking both NLS did not enter the nucleus, nuclear accumulation of HRG{alpha}1-241 is not diffusion dependent and requires at least one NLS, indicating an active nuclear import.

Subnuclear Dot Formation
We show here that HRG{alpha}1-241 exhibits a pattern of subnuclear dot formation similar to Sp100 and PML proteins (26) and similar to neuregulin-1 (20). The Sp100 and PML proteins were shown to be covalently modified by the SUMO-1 protein, which was partly required for dot formation (27). A consensus sequence for SUMOylation, (I/L)KXE, was proposed (28) and comparison of this consensus sequence with the sequences of both dot-forming domains in HRG{alpha}1-241 revealed that amino acids 28 to 52 contain an amino acid stretch very similar to the SUMOylation consensus sequence (data not shown). These data would support the findings of Golding et al., which show that the first 79 amino acids of NRG1ß3 were necessary and sufficient to direct the protein to nucleoli and nuclear speckles (20). However, we further showed that an additional domain (amino acids 113-148) is also required for dot formation. The nature of this second domain remains unclear because we were not able to identify a known consensus sequence in this domain. Thus, the mechanism of concerted action of both domains for dot formation remains to be elucidated. Nuclear dot-associated proteins were reported earlier to play a role in cell transformation and growth control or regulation of differentiation (29). Although NRG1ß3 was shown to associate with nucleoli and SC35-positive nuclear speckles, two nuclear compartments involved in ribosome synthesis, transcriptional control, and RNA splicing (20), the molecular composition and the biological function of the nuclear dots containing HRG{alpha}1-241 is not clear yet.

HRG{alpha}1-241 Interacts with Nuclear Proteins
At present, it is not known whether the mechanism underlying formation of HRG nuclear dots involves the post-translational modification of HRGs and/or the interaction of HRGs with other proteins. In our attempt to isolate genes encoding nuclear proteins interacting with HRG{alpha}1-241, we have screened a mammary gland cDNA library using a yeast two-hybrid system and obtained several candidate substrates (Table 1). This is the first study showing proteins interacting with HRG, which do not belong to the ErbB family of type I receptor tyrosine kinases.

Cullin-1 is a member of the SCF protein complex, an E3 ubiquitin protein ligase controlling the G1-S transition of the eukaryotic cell cycle (30), being expressed in the cytoplasm as well as in nuclear dots during interphase. Constitutive activation of nuclear factor-{kappa}B is observed in several cancers, including breast cancer. It was shown that HRG, but not its associated receptor ErbB2, plays a major role in constitutive nuclear factor-{kappa}B activation (31), thereby increasing the expression of proinvasive, prometastatic, and antiapoptotic genes in cancer cells, which leads to invasive and drug-resistant growth of breast cancer. Activation of nuclear factor-{kappa}B requires the degradation of its inhibitor I{kappa}B{alpha}, which is catalyzed by the SCF complex containing Cullin-1 (32). It is tempting to speculate that a direct interaction of intracellularly expressed HRG with Cullin-1 might circumvent the need for ErbB-dependent nuclear factor-{kappa}B activation.

MDGI was shown to display 95% homology with the heart fatty acid–binding protein (33), a member of a family of proteins thought to be involved in cell signaling, growth inhibition, and differentiation (34). MDGI is a nuclear protein that is maximally expressed in terminally differentiated mammary tissue (35) and has inhibitory activity on the growth of different breast cancer cell lines (36), reducing the transcriptional expression of c-fos, c-myc, and c-ras and suppressing the mitogenic effects of EGF (37). Nuclear translocation of HRGß1 is correlated with c-myc induction (13); however, further investigations are needed to understand the functional effect of the protein interaction between MDGI and HRG{alpha}1-241 on growth response and cell differentiation.

The G/T mismatch-specific TDG was originally cloned as a base excision repair enzyme (38) but was also shown to act as an activator (39) or repressor of transcriptional activity (40). It is not clear yet what the functional implication of the protein interaction of HRG{alpha}1-241 with TDG may be. However, we have found that HRG{alpha}1-241 has a potential role in transcriptional control supporting the physical interaction with TDG.

Another nuclear protein found to be a binding partner for HRG{alpha}1-241 is hUBC9, a protein showing significant identity with ubiquitin-conjugating enzymes required for cell cycle progression (41). The structure of hUBC9, however, displays significant differences with other ubiquitin-conjugating enzymes, which reflects its specificity for SUMO1 rather than for ubiquitin (42). In contrast to ubiquitination, SUMOylation does not tag proteins for degradation but seems rather to enhance their stability or modulate their subcellular compartmentalization (43). Furthermore, SUMOylation was correlated to transcriptional regulation (44) as well as altering protein activity and protein-protein interactions (45). As discussed above, HRG indeed contains a putative SUMOylation site; however, it remains to be defined if HRG{alpha}1-241 is indeed SUMOylated by hUBC9 and if this potential SUMOylation is necessary for the nuclear import, the formation of subnuclear dots, and/or the transcriptional activity of HRG{alpha}1-241.

To further specify these protein interactions, we constructed different HRG deletion variants and confirmed the ability of the specific domain(s) of HRG to interact with the nuclear proteins identified. Whereas Cullin-1 and MDGI showed very specific interaction with the EGF-like domain of HRG{alpha}1-24, suggesting that the interaction is probably not confined to the nucleus but instead plays a functional role in the cytoplasm, G/T mismatch-specific TDG was able to interact specifically with the Ig-like domain of HRG{alpha}1-241. However, MDGI and G/T mismatch-specific TDG showed activation of the reporter gene with all different deletion variants of HRG{alpha}1-241 as well as when transformed together with the empty bait vector. Because the activation of ß-galactosidase was significantly stronger when these proteins were cotransformed together with HRG{alpha}1-241, we assume that reporter gene activation by the other constructs may reflect background activation of the reporter gene and that the interaction with HRG{alpha}1-241 is specific and not due to an artifact resulting from the yeast two-hybrid system. Although hUBC9 was not able to activate the reporter gene without an interaction with HRG{alpha}1-241, this clone seemed to interact with both the Ig-like domain and the EGF-like domain. Interestingly, the interaction strength was much higher when both dot-forming sequences were present (HRG{alpha}1-241, minimal sequence), including the putative SUMOylation site. We assume that different parts of HRG{alpha}1-241 are necessary for the interaction with hUBC9, which act synergistically to enhance the protein-protein interaction.

Performing a coimmunoprecipitation assay based on transient overexpression of proteins in Cos-7 cells supported the findings from the yeast two-hybrid screen, whereby HRG{alpha}1-241 was interacting specifically with Cullin-1, MDGI, and G/T mismatch-specific TDG. However, we could not confirm the interaction of hUBC9 with HRG{alpha}1-241 with this approach. The reason might be that other factors are required for the formation of this complex or that the interaction of these two proteins must be triggered by an external signal. Another possibility might be that other intracellular substrates are titrating hUBC9 away from HRG{alpha}1-241.

HRGs Are Able to Form Dimers and Exhibit Transcriptional Repression Activity
Investigation of the dimerization potential is an important prerequisite for understanding how nuclear dots can be formed, because some proteins being part of nuclear dots have been shown to form homodimers. Dimerization can occur by means of a specific domain having the capability to self-interact and target the protein to discrete nuclear substructures (28). However, Sp100 has been shown to contain a self-aggregation domain that exists as separate entity besides the domain responsible for dot formation (46). To test the homomeric interaction potential of HRG{alpha}1-241, we have used the yeast two-hybrid assay and did coimmunoprecipitation experiments. Our results indicated that the minimal sequence shown to be sufficient for subnuclear dot formation of HRG{alpha}1-241 is capable of mediating HRG dimerization. The EGF-like domain, in contrast, completely lacked dimerization activity even when fused to a NLS. We do not know yet if different domains of HRG{alpha}1-241 are implicated in dimerization and subnuclear dot formation or if the same domains are responsible for both functions. Further mutational analyses have to be done to address this and further define the role of dimerization on HRG function.

Several proteins shown to form nuclear dots or to dimerize have either transcriptional transactivating or transrepressing properties (47, 48) or are implicated in DNA repair (49) or RNA splicing (50). HRG has been described to inhibit estrogen receptor expression (51) and to modulate the activity of c-myc after nuclear translocation (13). Because no DNA-binding site has been described for HRG{alpha}1-241 thus far, HRG{alpha}1-241 may regulate transcription indirectly by recruiting cofactors essential for transcriptional control. Indeed, in the yeast two-hybrid screen, we found that HRG{alpha}1-241 is specifically interacting with several proteins implicated in transcriptional regulation (HDAC2, MDGI, G/T DNA mismatch glycosylase, ZNF237, serine/arginine nuclear matrix protein, RING1 and YY1 binding protein, and p53-binding protein). Looking in more detail at the interaction of HRG with HDAC2, we were able to show interaction of endogenous HRG with HDAC2 in MDA-MB-231 cells. These data strongly support the hypothesis that nuclear HRG may be implicated in transcriptional control.

To assess the potential role of HRG{alpha}1-241 in transcriptional regulation, we devised a cellular luciferase reporter gene assay using two different constructs of the luciferase reporter gene either under the control of a constitutively active GAL4 promoter [pGAL4(5x UAS)-SV40-Luc] or under the control of a minimal GAL4 promoter (pGK-1). When Cos-7 cells were cotransfected with GAL4-DBD-HRG{alpha}1-241 together with either of the reporter gene constructs, we could show that HRG{alpha}1-241 was able to repress transcription when targeted to the promoter site of these constructs. When transfecting HDAC2 together with the reporter gene plasmids, it was able to repress luciferase expression; however, it failed to show any further transcriptional repression when cotransformed together with HRG{alpha}. These results suggest that the transcriptional repression by HRG{alpha} alone may already be maximal because of recruitment of endogenous cofactors by HRG{alpha}. At the moment, it is unclear which target genes may be regulated by HRG, but clearly testing for this activity against physiologic target reporter genes would be very informative.

In conclusion, we show nuclear localization of endogenous HRG in breast cancer cell lines and define the sequences necessary and sufficient for nuclear translocation and subsequent subnuclear dot formation by investigating the subcellular localization of transfected HRG{alpha}1-241. In our attempt to search for functional aspects of intracellular HRG{alpha}1-241, we show interaction of HRG{alpha}1-241 with different nuclear proteins and that these interactions are domain specific, either involving the Ig-like domain or the EGF-like domain or both. Furthermore, we report the ability of HRG proteins to form homodimers and reveal the HRG sequence responsible for dimerization. Additionally, we show interaction of endogenous HRG with HDAC2 and transcriptional regulation activity of HRG in a reporter gene assay. This is the first study showing interaction of HRG with nuclear proteins and our data clearly support a nuclear function of HRG in tumorigenesis.


   Materials and Methods
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
Acknowledgements
 References
 
Cell Lines
The hormone-dependent breast cancer cell line MCF-7 was obtained from Mason Research Institute (Rockville, MD) and grown in IMEM-ZO as described (52). Cos-7 cells were cultured in DMEM supplemented with 10% FCS. The breast cancer cell lines SKBR3 and MDA-MB-231 were obtained from American Type Culture Collection (Rockville, MD). T47D and BT474 cells were obtained from Dr. Nancy Hynes (Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland).

Endogenous Nuclear HRG Expression
Nuclear cell extracts were prepared using the CelLytic nuclear extraction kit (Sigma, St. Louis, MO) according to the recommendations of the manufacturer. Briefly, 107 cells were swelled in hypotonic lysis buffer followed by mechanical disruption. The cytoplasmic fraction was removed and the nuclear proteins were released from the nuclei by high-salt buffer. Western blots were done according to standard enhanced chemiluminescence procedures (Amersham, Otelfingen, Switzerland) using polyclonal anti-HRG antibodies (sc-347 and sc-348; Santa Cruz Biotechnologies, Santa Cruz, CA). Nuclear and cytoplasmic protein fractions were confirmed using antibodies against poly(ADP-ribose) polymerase (Cell Signaling Technologies, Beverly, MA) and MEK-1 (Zymed Laboratories, San Francisco, CA), respectively.

HRG{alpha} Vector Constructs
Total RNA was isolated from MDA-MB-231 cells with the RNeasy Total RNA kit (Qiagen, Basel, Switzerland), primed with oligo(dT) primers, and reverse transcribed with the First-Strand cDNA Synthesis kit (Clontech, Mountain View, CA). This reaction mixture was used as a template to amplify the extracellular part of HRG{alpha} (HRG{alpha}1-241). The resulting PCR product was cloned into the bacterial expression vector pCR2 (Invitrogen, Basel, Switzerland) by TA cloning and further subcloned into pEGFP-C1 (Clontech) to obtain a GFP fusion protein.

Construction of HRG{alpha} Deletion Mutants
COOH- and NH2-terminal sequential HRG{alpha}1-241 deletion mutants were constructed. The restriction enzymes used are listed from the 5' to the 3' end, SacII, SpeI, BclI, BbsI, and XmnI and, for the COOH-terminal deletion, XhoI. The COOH-terminal HRG{alpha}1-241 deletion constructs were obtained by linearization of pEGFP/HRG{alpha}1-241 at the appropriate sites, fill-in with T4 DNA polymerase, SmaI digestion in the multiple cloning site of pEGFP-C1 3' to the HRG{alpha}1-241 insert, and religation. The NH2-terminal HRG{alpha}1-241 deletions were constructed to be in-frame with the GFP coding sequence after digestion, fill-in, and religation. The pEGFP/NLS plasmid was obtained by ligating a phosphorylated synthetic oligonucleotide linker (pN1 3'-TCGATATCCAAAGAAGAAGCGCAAGGTGCA-5' and pN2 3'-CCTTGCGCTTCTTCTTTGGATA-5') into pEGFP-C1 digested with XhoI and PstI.

Transient Expression of HRG Constructs in Mammalian Cells
MCF-7 cells were used for transient expression analysis of HRG{alpha}1-241 constructs. Transfections were carried out by electroporation with the Gene Pulser II (Bio-Rad Technologies, Reinach, Switzerland). Alternatively, MCF-7 cells were grown on glass coverslips to 60% to 80% confluence and transfected for 3 hours with 1 µg plasmid DNA using Superfect reagent (Qiagen). For coimmunoprecipitation studies, Cos-7 cells were transfected with Fugene 6 transfection reagent (Roche Diagnostics, Rotkreuz, Switzerland).

Fluorescence Microscopy
Transfected cells were seeded on glass coverslips in six-well plates and incubated for 24 hours. Cell nuclei transfected with GFP fusion constructs were stained with either 2 µg/mL Hoechst 33342 dye (blue) or 5 µmol/L SYTO 17 red fluorescent nucleic acid stain (Molecular Probes, Eugene, OR). Alternatively, the cells were fixed in ice-cold methanol/acetone (1:1, v/v) at –20°C and the coverslips were allowed to dry. Coverslips were subsequently mounted with 90% glycerol in 1x PBS onto microscope slides. Cells were observed with a Zeiss Axioskop microscope. Image acquisition was done with a CCD camera using the MacProbe program (Perceptive Scientific Instruments Ltd., Suffolk, England).

Yeast Two-Hybrid Screen
The yeast two-hybrid assay was done using the Matchmaker cloning system 3 according to the recommendations of the manufacturer (Clontech). HRG{alpha}1-241 was subcloned into pGBKT7 vector in-frame with the GAL4-DBD. This fusion construct was used to screen, on histidine-free medium, a human mammary gland cDNA library (Clontech) cloned into the pACT2 vector in-frame with the GAL4 activation domain. The GAL4-responsive AH109 yeast strain was used for the screening and transformation was accomplished according to the lithium acetate transformation protocol. Positive colonies were reselected on adenine-free medium and the relative strength of the interactions was assessed using a liquid ß-galactosidase assay. Positive clones were rescued via transformation of DH5{alpha} bacteria and subsequent selection with ampicillin and analyzed by DNA sequencing. To evaluate the interaction of the positive clones with different domains of HRG{alpha}1-241, the individual HRG deletion variants coding for the Ig-like domain (amino acids 1-128), the EGF-like domain (amino acids 121-241), or the minimal sequence (amino acids 21-150), necessary and sufficient for nuclear translocation and subnuclear dot formation, have been cloned into the pGBKT7 and the pGADT7 vector.

Coimmunoprecipitation
The myc epitope tag and the HA epitope tag were subcloned into the pcDNA3.1(+) vector (Invitrogen). HRG{alpha}1-241, its deletion variants, and the full-length cDNA of some positive clones were subsequently subcloned into the pcDNA3.1/myc and the pcDNA3.1/HA vectors, respectively. Cos-7 cells were transiently cotransfected with pcDNA3.1/myc containing a HRG variant together with pcDNA3.1/HA containing either the full-length cDNA of a positive clone or a HRG variant. All transfections included 0.2 µg p6RlacZ vector (53) to measure transfection efficiency and to define the input amount of protein for each immunoprecipitation. For immunoprecipitation, cells were lysed 48 hours after transfection in NP40 lysis buffer (150 mmol/L NaCl, 1% NP40, 50 mmol/L Tris, 1 mmol/L EDTA) supplemented with protease inhibitors. Cell lysates were incubated overnight with protein G-Sepharose 4 fast flow (Amersham) preincubated with a monoclonal anti-myc antibody 9B11 (Cell Signaling). The supernatant was precipitated using TCA and protein G-Sepharose was washed in immunoprecipitation buffer [20 mmol/L K+ HEPES (pH 7.4), 200 mmol/L sucrose, 1 mmol/L EDTA, 100 mmol/L NaCl] supplemented with protease inhibitors. Immunoprecipitated material was solubilized by resuspending the washed beads in 2x SDS protein sample buffer containing ß-mercaptoethanol. The immunoprecipitation efficiency was ~50% to 80%. Western blots were done according to standard enhanced chemiluminescence procedures (Amersham) using a polyclonal anti-HA antibody Y-11 (Santa Cruz). For coimmunoprecipitation of endogenous HRG, cells were lysed in extraction buffer (50 mmol/L HEPES, 150 mmol/L NaCl, 25 mmol/L ß-glycerophosphate, 25 mmol/L NaF, 5 mmol/L EGTA, 1 mmol/L EDTA, 15 mmol/L PPI) without detergent, supplemented with protease inhibitors. Cell lysates were immunoprecipitated using a polyclonal anti-HRG antibody (Santa Cruz), and Western blots were done using a monoclonal anti-HDAC2 antibody (Upstate, Lake Placid, NY).

Luciferase Reporter Gene Activation Assay
Cos-7 cells were seeded into six-well plates 24 hours before transfection. All transfections included 0.2 µg p6RlacZ vector (53) for normalization of transfection efficiency. Standard amounts of expression and reporter plasmids per transfection in reporter gene activation assays were 1 µg GAL4-DBD expression vector pcDNA3.1/GAL4-DBD or 1 µg HRG expression plasmid pcDNA3.1/GAL4-DBD-HRG{alpha}1-241, 1 µg GAL4-responsive luciferase reporter plasmid pGAL4(5x UAS)-SV40-Luc (under the control of a constitutively active GAL4 promoter), or pGK-1 (under the control of a minimal GAL4 promoter; ref. 54) and, optionally, 1 µg pcDNA3.1/HA/HDAC2. Cell lysates were prepared 48 hours after transfection and subsequently assayed for luciferase and ß-galactosidase activity. Luciferase values normalized to ß-galactosidase activity are called relative luciferase units. The data shown represent mean ± SD of three independent experiments done.


   Acknowledgements
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
References
 
We thank Gerhard Christofori, Heidi Lane, and Chris Benz for their mutual interest and support in this work; Wilhelm Krek for the hUBC9 and Cullin-1 expression vectors; Nancy Hynes for the T47D and BT474 cells; Natasha Kralli for the pGK-1 and pGAL4(5xUAS)-SV40-Luc plasmids; and Francois David and Heidi Bodmer for their technical assistance.


   Notes
Top
 Notes
Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Grant support: Swiss National Science Foundation grants SNF 3100-059819.99 and SNF 3100-49505.96 (U. Eppenberger).

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. Breuleux is currently at Novartis Institutes for Biomedical Research, Oncology, Basel, Switzerland. F. Schoumacher and D. Rehn are currently at F. Hoffmann-La Roche AG, Basel, Switzerland. W. Küng is currently at the Department of Research, Medical Oncology, University Hospital Basel, Basel, Switzerland. H. Mueller is currently at the Institute of Biochemistry and Genetics, University of Basel, Basel, Switzerland. U. Eppenberger is currently at Stiftung Tumorbank Basel, Riehen, Switzerland.

Received 2/16/05; revised 11/21/05; accepted 12/19/05.


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

  1. Wen D, Suggs SV, Karunagaran D, et al. Structural and functional aspects of the multiplicity of Neu differentiation factors. Mol Cell Biol 1994;14:1909–19.[Abstract/Free Full Text]
  2. Carraway KlR, Weber JL, Unger MJ, et al. Neuregulin-2, a new ligand of ErbB3/ErbB4-receptor tyrosine kinases. Nature 1997;387:512–6.[CrossRef][Medline]
  3. Harari D, Tzahar E, Romano J, et al. Neuregulin-4: a novel growth factor that acts through the ErbB-4 receptor tyrosine kinase. Oncogene 1999;18:2681–9.[CrossRef][Medline]
  4. Zhang D, Sliwkowski MX, Mark M, et al. Neuregulin-3 (NRG3): a novel neural tissue-enriched protein that binds and activates ErbB4. Proc Natl Acad Sci U S A 1997;94:9562–7.[Abstract/Free Full Text]
  5. Burgess TL, Ross SL, Qian YX, Brankow D, Hu S. Biosynthetic processing of neu differentiation factor. Glycosylation trafficking, and regulated cleavage from the cell surface. J Biol Chem 1995;270:19188–96.[Abstract/Free Full Text]
  6. Loeb JA, Fischbach GD. ARIA can be released from extracellular matrix through cleavage of a heparin-binding domain. J Cell Biol 1995;130:127–35.[Abstract/Free Full Text]
  7. Barbacci EG, Guarino BC, Stroh JG, et al. The structural basis for the specificity of epidermal growth factor and heregulin binding. J Biol Chem 1995;270:9585–9.[Abstract/Free Full Text]
  8. Holmes WE, Sliwkowski MX, Akita RW, et al. Identification of heregulin, a specific activator of p185erbB2. Science 1992;256:1205–10.[Abstract/Free Full Text]
  9. Alroy I, Yarden Y. The ErbB signaling network in embryogenesis and oncogenesis: signal diversification through combinatorial ligand-receptor interactions. FEBS Lett 1997;410:83–6.[CrossRef][Medline]
  10. Yen L, You XL, Al Moustafa AE, et al. Heregulin selectively upregulates vascular endothelial growth factor secretion in cancer cells and stimulates angiogenesis. Oncogene 2000;19:3460–9.[CrossRef][Medline]
  11. Atlas E, Cardillo M, Mehmi I, Zahedkargaran H, Tang C, Lupu R. Heregulin is sufficient for the promotion of tumorigenicity and metastasis of breast cancer cells in vivo. Mol Cancer Res 2003;1:165–75.[Abstract/Free Full Text]
  12. Tsai MS, Shamon-Taylor LA, Mehmi I, Tang CK, Lupu R. Blockage of heregulin expression inhibits tumorigenicity and metastasis of breast cancer. Oncogene 2003;22:761–8.[CrossRef][Medline]
  13. Li W, Park JW, Nuijens A, Sliwkowski MX, Keller GA. Heregulin is rapidly translocated to the nucleus and its transport is correlated with c-myc induction in breast cancer cells. Oncogene 1996;12:2473–7.[Medline]
  14. Fluge O, Akslen LA, Haugen D, Varhaug JE, Lillehaug JR. Expression of heregulins and associations with the ErbB family of tyrosine kinase receptors in papillary thyroid carcinomas. Int J Cancer 2000;87:763–70.[Medline]
  15. Gilbertson RJ, Clifford SC, MacMeekin, et al. Expression of the ErbB-neuregulin signaling network during human cerebellar development: implications for the biology of medulloblastoma. Cancer Res 1998;58:3932–42.[Abstract/Free Full Text]
  16. Marshall C, Blackburn E, Clark M, Humphreys S, Gullick WJ. Neuregulins 1-4 are expressed in the cytoplasm or nuclei of ductal carcinoma (in situ) of the human breast. Breast Cancer Res Treat 2005;27:1–6.
  17. Kimura H. Schwannoma-derived growth factor must be transported into the nucleus to exert its mitogenic activity. Proc Natl Acad Sci U S A 1993;90:2165–9.[Abstract/Free Full Text]
  18. Lin YZ, Yao SY, Hawiger J. Role of the nuclear localization sequence in fibroblast growth factor-1-stimulated mitogenic pathways. J Biol Chem 1996;271:5305–8.[Abstract/Free Full Text]
  19. Adam RM, Danciu T, Mclellan DL, et al. A nuclear form of the heparin-binding epidermal growth factor-like growth factor precursor is a feature of aggressive transitional cell carcinoma. Cancer Res 2003;63:484–90.[Abstract/Free Full Text]
  20. Golding M, Ruhrberg C, Sandle J, Gullick WJ. Mapping nucleolar and spliceosome localization sequences of neuregulin1-3. Exp Cell Res 2004;299:110–8.[CrossRef][Medline]
  21. Hicks GR, Raikhel NV. Protein import into the nucleus: an integrated view. Annu Rev Cell Dev Biol 1995;11:155–88.[CrossRef][Medline]
  22. Fields S, Song O. A novel genetic system to detect protein-protein interactions. Nature 1989;340:245–6.[CrossRef][Medline]
  23. Cress WD, Seto EJ. Histone deacetylases, transcriptional control and cancer. J Cell Physiol 2000;184:1–16.[CrossRef][Medline]
  24. Johnson HM, Subramaniam PS, Olsnes S, Jans DA. Trafficking and signaling pathways of nuclear localizing protein ligands and their receptors. Bioessays 2004;26:993–1004.[CrossRef][Medline]
  25. Chelsky D, Ralph R, Jonak G. Sequence requirements for synthetic peptide-mediated translocation to the nucleus. Mol Cell Biol 1989;9:2487–92.[Abstract/Free Full Text]
  26. Sternsdorf T, Jensen K, Will H. Evidence for covalent modification of the nuclear dot-associated proteins PML and Sp100 by PIC1/SUMO-1. J Cell Biol 1997;139:1621–34.[Abstract/Free Full Text]
  27. Müller S, Matunis MJ, Dejean A. Conjugation with the ubiquitin-related modifier SUMO-1 regulates the partitioning of PML within the nucleus. EMBO J 1998;17:61–70.[CrossRef][Medline]
  28. Sternsdorf T, Jensen K, Reich B, Will H. The nuclear dot protein Sp100, characterization of domains necessary for dimerization, subcellular localization, and modification by small ubiquitin-like modifiers. J Biol Chem 1999;274:12555–66.[Abstract/Free Full Text]
  29. Le XF, Yang P, Chang KS. Analysis of the growth and transformation suppressor domains of promyelocytic leukemia gene, PML. J Biol Chem 1996;271:130–5.[Abstract/Free Full Text]
  30. Krek W. Proteolysis and the G1-S transition: the SCF connection. Curr Opin Genet Dev 1998;8:36–42.[CrossRef][Medline]
  31. Bhat-Nakshatri P, Sweeney CJ, Nakshatri H. Identification of signal transduction pathways involved in constitutive NF-{kappa}B activation in breast cancer cells. Oncogene 2002;21:2066–78.[CrossRef][Medline]
  32. Tan P, Fuchs SY, Chen A, et al. Recruitment of a ROC1-1 ubiquitin ligase by Skp1 and HOS to catalyze the ubiquitination of I{kappa}B{alpha}. Mol Cell 1999;3:527–33.[CrossRef][Medline]
  33. Spener F, Unterberg C, Borchers T, Grosse R. Characteristics of fatty acid-binding proteins and their relation to mammary-derived growth inhibitor. Mol Cell Biochem 1990;98:57–68.[Medline]
  34. Borchers T, Hohoff C, Buhlmann C, Spener F. Heart-type fatty acid binding protein—involvement in growth inhibition and differentiation. Prostaglandins Leukot Essent Fatty Acids 1997;57:77–84.[CrossRef][Medline]
  35. Politis