Intersectin 1 Enhances Cbl Ubiquitylation of Epidermal Growth Factor Receptor through Regulation of Sprouty2-Cbl Interaction

Abstract

Ubiquitylation of receptor tyrosine kinases plays a critical role in regulating the trafficking and lysosomal degradation of these important signaling molecules. We identified the multidomain scaffolding protein intersectin 1 (ITSN1) as an important regulator of this process (N. P. Martin et al., Mol. Pharmacol. 70:1463–1653, 2006) ITSN1 stimulates ubiquitylation of the epidermal growth factor receptor (EGFR) through enhancing the activity of the Cbl E3 ubiquitin ligase. However, the precise mechanism through which ITSN1 enhances Cbl activity was unclear. In this study, we found that ITSN1 enhances Cbl activity through disrupting the interaction of Cbl with the Sprouty2 (Spry2) inhibitory protein. We demonstrate that ITSN1 binds Pro-rich regions in both Cbl and Spry2 and that interaction of ITSN1 with Spry2 disrupts Spry2-Cbl interaction, resulting in enhanced ubiquitylation of the EGFR. Disruption of ITSN1 binding to Spry2 through point mutation of the Pro-rich ITSN1 binding site in Spry2 results in enhanced Cbl-Spry2 interaction and inhibition of receptor ubiquitylation. This study demonstrates that ITSN1 enhances Cbl activity by modulating the interaction of Cbl with Spry2. In addition, our results reveal a new level of complexity in the regulation of Cbl through the interaction with ITSN1 and Spry2.

INTRODUCTION

Receptor tyrosine kinases (RTKs) play critical roles in the regulation of multiple aspects of metazoan life. Binding of ligand stimulates the intrinsic kinase activity of the receptor, leading to the recruitment and activation of numerous intracellular signaling pathways. However, a number of mechanisms exist to regulate the extent and duration of RTK signaling. One such mechanism involves the covalent attachment of ubiquitin to activated receptors. This posttranslational modification targets the activated receptors for lysosomal degradation (19). Thus, regulation of RTK ubiquitylation represents a critical step in cellular signaling.

Cbl is a RING (really interesting new gene) domain E3 ubiquitin ligase that specifically regulates RTK ubiquitylation (28). Although binding of Cbl to activated RTKs represents an important step in regulation of RTK ubiquitylation, Cbl activity is modulated by both posttranslational modifications as well as interactions with numerous proteins (28). One such protein is the intersectin 1 (ITSN1) scaffold protein. Although initially identified as a regulator of clathrin-dependent endocytosis, ITSN1 regulates a number of additional biochemical pathways (25). Recently, we demonstrated that ITSN1 enhances Cbl-dependent ubiquitylation of the epidermal growth factor receptor (EGFR), leading to enhanced degradation of the activated receptor (20). However, the mechanism underlying the increase in Cbl activity was unclear. We postulated that ITSN1 either promoted Cbl binding to an activator or prevented Cbl interaction with a negative regulator. In this study, we defined a novel role for ITSN1 in attenuating Cbl inhibition by Spry2, a negative regulator of Cbl (9, 15). Our results demonstrate that ITSN1 binds both Cbl and Spry2 and that ITSN1 releases Cbl from Spry2 inhibition, leading to enhanced EGFR ubiquitylation.

MATERIALS AND METHODS

Cell lines and reagents.

HEK293T human kidney epithelial cells and COS-1 monkey kidney cells were maintained in Dulbecco's modified Eagle medium (DMEM) with 10% fetal bovine serum. Human IMR-5 neuroblastoma cells were grown in RPMI medium supplemented with 10% fetal bovine serum. All cells were grown at 37°C in a humidified chamber with 5% CO₂–95% air. Epidermal growth factor was purchased from Millipore. The antibodies used in this study were N-Spry2 and ubiquitin P4D1 antibodies (Santa Cruz), EGFR AB12 and EGFR AB13 antibodies (Thermo Scientific), and monoclonal antihemagglutinin (HA) antibody (Covance).

DNA constructs and transfection.

An amino-terminal HA epitope-tagged full-length ITSN1 (mouse) in pCGN construct was previously described (24). HA-tagged wild-type (WT) human c-Cbl was a gift from Yosef Yarden (Weizmann Institute of Science, Rehovot, Israel) and was described previously (18). The pHM6-HA-Spry2 and its empty vector, pHM6-HA, were kindly provided by Tarun Patel (Loyola University, Chicago, IL) and were described previously (38). COS-1 cells were transfected with Lipofectamine (Invitrogen, Carlsbad, CA) according to the protocol provided by manufacturer. Glutathione S-transferase (GST)-tagged SH3 domains of ITSN were created by subcloning the individual SH3 domains into the mammalian expression vector pEFG (26). The Spry2 mutants containing single-amino-acid mutations (Y55F, P59A, P65A, P69A, P71A, P73A, P304A, and P308A) were generated from the plasmid pCEFL-KZ-AU5-Spry2 WT (4, 22) by site-directed PCR mutagenesis using specific primers. The sequences of all PCR-generated constructs were verified by direct sequencing, and those of the oligonucleotides used are available upon request. Spry2 wild-type (WT), Y55F, P59A, and P308A fragments were subcloned into pHA-VC155, kindly provided by Chang-Deng Hu (Purdue University, West Lafayette, IN).

COOH-terminal truncated constructs of Spry2 from amino acid 301 (T301) in pXJ40-FLAG have been described (17). Spry2N and Spry2C were also previously described (2). Various truncation mutants of the short isoform of ITSN1 were generated using the reverse primer 5′CGGGGTACCCCGAGATGCAGGTCTGAGCACC3′ and the following forward primers: ΔEH1, 5′ATAAGAATGCGGCCGCTGTCATGAAACAGGCAACCAGTG3′; ΔEH1+EH2, 5′ATAAGAATGCGGCCGCTCAGCCACTGCCGCCCGTC3′; and ΔEH1+EH2+CC, 5′ATAAGAATGCGGCCGCTCATCAGGAGCCAGCTAAGCTG3′. The N-terminal truncation mutants were cloned into pXJ40-Myc using NotI and KpnI sites.

Immunoprecipitation and immunoblotting.

Whole-cell extracts were prepared as described previously (26). For the analysis of endogenous levels of ubiquitin in COS-1 cells, lysis buffer was supplemented with 5 mM N-ethylmaleimide. EGFR immunoprecipitation and ubiquitylation levels were determined as previously described (20). For detection of Spry2, EGFR, and HA-tagged proteins, standard protocols suggested by the manufacturers were used.

GST pulldown assays.

Samples were lysed in a Tris-based buffer (20 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton, 2.5 mM sodium pyrophosphate, 1 mM sodium orthovanadate, and a cocktail of protease inhibitors) and centrifuged at 13,200 rpm for 15 min at 4°C. Fifteen microliters of glutathione Sepharose4B beads (Amersham Biosciences, Buckinghamshire, United Kingdom) was added to the supernatant to precipitate the GST epitope. The resulting immunoprecipitates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Yeast two-hybrid screening.

Yeast two-hybrid screening analysis was performed through a contract with Myriad Genetics essentially as described previously (3, 33) except that the various individual domains of mouse or human ITSN1 were used as bait. Multiple mouse and human Spry2 clones were identified as binding to the first SH3 domain of ITSN1.

Peptide screening of SH3 domain blots.

SH3 domain blots were obtained from Panomics, Inc. (Redwood City, CA). Biotinylated peptides (1 μg/ml) coding for the Pro-rich region of Spry2, TVCCKVPTVPPRNFEKPT, or a control peptide, TVCCKVATVPANFEKPT, were incubated with membranes overnight at 4°C. The membranes were washed with phosphate-buffered saline (PBS) containing 0.01% Tween 20 (PBST) three times for 15 min each time and then incubated with streptavidin-conjugated horseradish peroxidase (1:100,000 in PBST). After three 15-min washes with PBST, an enhanced chemiluminescent system (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom) was used to detect bound peptides.

Bimolecular fluorescence complementation (BiFC).

BiFC was performed essentially as described previously (36). Briefly, COS-1 cells were seeded on glass-bottom plates and in 6-well dishes and transfected with 0.5 μg of plasmids encoding proteins fused to pHA-VC155N and pFLAG-VN173N. At 24 h posttransfection, the glass bottom dishes were fixed on ice in 3.7% formaldehyde for 20 min, rinsed twice with PBS, and stored in PBS at 4°C in the dark. A Zeiss LSM 510 META confocal microscope was used to image samples. Cyan fluorescent protein (CFP)-positive cells were selected and imaged for BiFC signal in the yellow fluorescent protein channel. BiFC was quantified and expressed as average fluorescence intensity per pixel using ImageJ, available from the National Institutes of Health (NIH) as described previously (36). In parallel with imaging, cells in 6-well dishes were lysed and expression levels of transfected proteins were determined by Western blot analysis.

RESULTS

Spry2 is an ITSN binding partner.

Spry2 was identified in a high-throughput yeast two-hybrid (Y2H) screen designed to identify ITSN-binding partners (K. A. Wong et al., unpublished observations). The SH3A domain of ITSN1 (amino acids [aa] 730 to 816) isolated both human and mouse Spry2 clones as targets. The COOH terminus of Spry2 contains a consensus Pro-rich sequence (PTVPPRN) resembling the ligand for ITSN1's first SH3 domain, SH3A (3, 20). Using a biotinylated peptide derived from the Spry2 sequence encompassing this site (TVCCKVPTVPPRNFEKPT), we identified the SH3 domains of both ITSN1 and ITSN2 as potential binding partners for Spry2. Thus, both Y2H and peptide screening experiments suggest that ITSN1 and Spry2 may represent binding partners in vivo.

ITSN1 and Spry2 interact in cells.

Immunocytochemical staining of cells reveals that a portion of endogenous ITSN1 and Spry2 colocalizes in cells (Fig. 1A). Although we were unable to coprecipitate endogenous ITSN1 and Spry2 from cells, possibly because the antibodies target epitopes in the regions of interaction between the two proteins, we analyzed the interaction of epitope-tagged versions of the two proteins (Fig. 1B). Using HA-epitope tagged versions of the major ITSN1 isoforms (25), we demonstrated that both ITSN1-S and ITSN1-L interact with Spry2, suggesting that the presence of the guanine nucleotide exchange factor (GEF) domain on ITSN1-L does not interfere with Spry2 interaction (Fig. 1B). Spry2 is a member of the Spry family of proteins, which consists of Spry1 to Spry4 (15). To determine the specificity of ITSN1 for specific Spry members, we coexpressed ITSN1 with different Spry isoforms (Fig. 1C and D). ITSN1 specifically interacted with full-length Spry2, and this binding was abolished by deletion of the COOH-terminal Pro-rich tail in the Spry2 T301 truncation mutant. Spry4, which lacks a comparable Pro-rich sequence, did not interact with ITSN1 (Fig. 1C and D).

Fig 1.

Endogenous ITSN1 binds Spry2. (A) Endogenous ITSN1 (red) colocalizes with endogenous Spry2 (green) in IMR-5 neuroblastoma cells (top row). As controls (middle and bottom rows), cells were stained with both fluorescently labeled secondary antibodies (Cy5-labeled donkey anti-rabbit and fluorescein isothiocyanate (FITC)-labeled donkey anti-mouse) but only a single primary antibody, as indicated in each panel. (B) Spry2 binds both ITSN1 isoforms. FLAG-tagged Spry2 was coexpressed with either HA-tagged ITSN1-S or ITSN1-L. Both isoforms are detected in FLAG immunoprecipitates. (C) Schematic of Spry constructs. (D) HEK293T cells were transiently transfected with the constructs indicated above the gels. ITSN1 SH3A-E was epitope tagged with a Myc epitope, whereas the Spry2 constructs were tagged with FLAG. Proteins were immunoprecipitated with either Myc or FLAG antibodies, as indicated on the left. Western blots of the immunoprecipitates were then probed with the antibodies indicated on the right. Expression of the various proteins is indicated in the Western blots of cell lysates shown in the bottom two panels. N, Spry2N; C, Spry2C.

Using truncation mutants of ITSN1, we observed that ITSN1's SH3 domains mediated Spry2 binding (see Fig. 4). Given the presence of five SH3 domains (domains A to E) in ITSN1, we examined the specificity of Spry2 for each of these SH3 domains. The five SH3 domains were individually cloned into the mammalian expression vector pEFG (26) as described in Materials and Methods. These SH3 constructs were cotransfected into HEK293T cells along with FLAG-Spry2. Following immunoprecipitation with anti-FLAG antibody, we observed that the SH3A domain of ITSN1 but not any of the other SH3 domains specifically interacted with full-length Spry2 (Fig. 2A). Although Spry2 contains two Pro-rich stretches (aa 59-PTVVPRP-65 and aa 304-PTVPPRN-310), only mutation of Pro304 to Ala (P304A) in the COOH-terminal Pro-rich sequence disrupted binding of ITSN1 (Fig. 2B).

Fig 4.

The NH₂ terminus of ITSN1 negatively regulates binding to Spry2. (A) Schematic of ITSN1 NH₂-terminal-truncation mutants. (B) HEK293T cells were cotransfected with FLAG-tagged wild type Spry2 along with full-length ITSN1-S or ITSN1 with various NH₂-terminal truncations. (Top) FLAG-tagged Spry2 was immunoprecipitated using M2 beads to identify the ITSN1 truncations that interact with Spry2. (Middle) A 10×-longer exposure of the blot in the top panel. (Bottom) Levels of protein expression from the various transfected gene constructs were similar. (C) Immunoprecipitation of Pro-rich Spry2 peptides with various NH₂-terminal truncations of ITSN-S. HEK293T cells were transfected with full-length ITSN1 or ITSN1 with various NH₂-terminal truncations. Cell lysates were incubated with either biotinylated Pro-rich Spry2 peptides or control peptides. (Top) Biotin-labeled peptides were immunoprecipitated using streptavidin Sepharose beads. Precipitates were analyzed by immunoblotting with Myc antibodies to detect the ITSN1 proteins. (Bottom) Levels of expression of the various Myc-tagged ITSN1 proteins in cell lysates.

Fig 2.

Mapping Spry2-ITSN1 interactions. (A) GST-tagged constructs of the individual SH3 domains of ITSN1 were coexpressed in HEK293T cells along with FLAG-Spry2. Only the SH3A domain of ITSN1 coprecipitates with Spry2. Control (GST alone) is not visible in the cell lysate blot due to its smaller size. (B) Mutation of Pro304 disrupts ITSN1 binding. AU5-tagged Spry2 wild type (WT) or point mutants containing Pro-Ala substitutions at the indicated amino acids were coexpressed with GST-SH3A in HEK293T cells. Following purification of the SH3A domain from cell lysates using glutathione beads, Western blots were performed to detect association of Spry2 proteins. Mutation of P304A disrupted ITSN1 SH3A binding, whereas the other Pro mutations had little or no effect.

To examine the interaction of Spry2 and ITSN1 in whole cells, we utilized bimolecular fluorescence complementation (BiFC) (Fig. 3). As seen with the individual SH3A domain of ITSN1, Spry2 interaction with full-length ITSN1 was disrupted by the P304A mutation but not by the Y55A mutation (Fig. 3). Mutation of P59A resulted in slight but significant reduction in ITSN1 interaction. These differences in BiFC signal were not due to differences in expression of the various Spry2 mutants and thus likely reflect true differences in the affinity of ITSN1 for these mutants (Fig. 3C). These findings demonstrate that ITSN1 specifically interacts with the COOH-terminal Pro-rich sequence in Spry2.

Fig 3.

Interaction of Spry2 and ITSN1 by BiFC. (A) VC-tagged Spry2 WT or various Spry2 mutants were coexpressed with VN-ITSN1 in COS cells. CFP was included at one-fifth the amount of DNA as a transfection control. pep, a nonspecific peptide control fused to VC. (B) Interaction of Spry2 and ITSN1 was quantified as described previously (36). WT, Y55A, and P59A Spry2 proteins interacted with VN-ITSN1, whereas P304A Spry2 was impaired in the interaction. Experiments were performed in duplicate. Data are average fluorescence intensities per cell; error bars show standard errors of the means (SEM). Asterisks indicate that the values for these Spry2 mutants were significantly different from that for wild-type Spry2 (P < 0.05). (C) A Western blot of lysates from the BiFC experiments demonstrates equivalent expression of the tagged proteins.

SH3 binding to targets is negatively regulated by ITSN1's EH and CC domains.

During the course of our investigations, we observed that Spry2 interacted better with the isolated SH3A domain than with full-length ITSN1 (data not shown). One possible explanation for these results is that the regions NH₂ terminal to the SH3 domains, i.e., the EH and CC domains, may sterically hinder SH3 binding to targets such as Spry2. To test this possibility, we created a series of ITSN1 mutants with NH₂-terminal truncations which were tested for interaction with Spry2 (Fig. 4A). Myc-tagged full-length or truncated ITSN1 proteins were coexpressed with Spry2 in HEK293T cells. Immunoprecipitation of Spry2 revealed increased binding to ITSN1 with progressive truncation of the NH₂ terminus (Fig. 4B). Deletion of the EH1 domain enhanced Spry2 binding to ITSN1 compared to its binding to full-length ITSN1. Although not visible on the gel in Fig. 4B, full-length ITSN1-S and ITSN1-L do indeed interact with Spry2 by coimmunoprecipitation (Fig. 1B). Removal of both EH domains of ITSN1 did not appear to further enhance binding to Spry2. However, deletion of the EH and CC domains further enhanced Spry2-ITSN1 interaction. Similar results were observed in the binding of another ITSN1 target, N-WASP, which also interacts with ITSN1's SH3 domains (data not shown). Although Spry2 bound exclusively to SH3A (Fig. 2A), N-WASP interacted with multiple SH3 domains (SH3A > SH3C > SH3E > SH3D). However, SH3B did not interact with N-WASP. These findings are consistent with previous reports demonstrating ITSN1 binding to N-WASP proteins (13, 40). To further confirm that the SH3 domains of ITSN1 are sterically hindered in the full-length protein and to circumvent the possibility that the Pro-rich motif of Spry2 or N-WASP may not be properly presented for binding, a biotinylated Pro-rich Spry2 peptide was used in a pulldown assay. Biotinylated peptides were incubated with cell lysates from HEK293T cells transfected with the various NH₂-terminal truncation mutant of the ITSN1 short isoform (ITSN1-S). The biotinylated peptides were preincubated with streptavidin-conjugated Sepharose beads and then mixed with cell lysates. Consistent with the results in Fig. 4B, we observed increased ITSN1 binding to the Spry2 peptide upon progressive NH₂-terminal truncations in ITSN1, with the isolated SH3 region binding most avidly to the biotinylated Spry2 peptide (Fig. 4C).

ITSN1 disrupts Spry2 interaction with Cbl to enhance EGFR ubiquitylation.

We previously demonstrated that ITSN1 regulates EGFR degradation through enhancing Cbl ubiquitylation of the activated EGFR (20). Since ITSN1 did not affect Cbl binding to EGFR, Cbl phosphorylation, or Cbl stability (20), we speculated that ITSN1 might activate Cbl by disrupting the interaction with Cbl inhibitory proteins. Thus, the identification of Spry2 (a Cbl inhibitor) as an ITSN1 binding partner suggests that ITSN1 might activate Cbl by disrupting the Spry2-Cbl interaction, leading to enhanced ubiquitylation of the EGFR. To test this possibility, we examined the effect of ITSN1 overexpression on Spry2-Cbl interaction and EGFR ubiquitylation. Using BiFC to quantify Spry2-Cbl binding, we observed that ITSN1 decreased Spry2-Cbl binding in a dose-dependent manner (Fig. 5A and B). The loss of Spry2-Cbl BiFC signal was not due to changes in the expression of VN-Spry2 or VC-Cbl (Fig. 5C). Using epitope-tagged versions of these proteins instead of BiFC, we also demonstrated that ITSN1 dose-dependently decreased the coimmunoprecipitation of Spry2 with Cbl, thus corroborating the BiFC data (Fig. 5D).

Fig 5.

ITSN1 disrupts Spry2-Cbl interaction. (A) Spry2-Cbl interaction was measured by BiFC. ITSN1 expression leads to a dose-dependent decrease in Spry2-Cbl interaction. Coexpression of VN-Spry2 with VC-pep, a nonspecific peptide control, does not result in a BiFC signal. (B) Quantification of BiFC signal. Interaction of Spry2 and Cbl was quantified as described previously (36). Results are the averages and SEM from three independent experiments. Asterisks indicate significant differences from the result for the VN-Spyr2+VC-Cbl sample (P < 0.05). (C) Western blotting demonstrates the expression of the various proteins. Both ITSN1 and Cbl were HA tagged. The differences in Spry2-Cbl interaction are not due to changes in the overall expression of these proteins. (D) Overexpression of HA epitope-tagged ITSN1 dose dependently disrupts the binding of Spry2 WT to Cbl. HA-Cbl was immunoprecipitated from cells and the coprecipitation of Spry2 was monitored by Western blot of Cbl precipitates. (Top) Western blots of anti-Cbl precipitates with the indicated antibodies. (Bottom) Western blots of cell lysates with the indicated antibodies.

Given the ability of ITSN1 to disrupt Spry2-Cbl interaction, we next tested the possibility that increasing ITSN1 levels might reverse Spry2 inhibition of Cbl. Transient overexpression of Cbl enhanced EGF-stimulated ubiquitylation of endogenous EGFR, and coexpression of Spry2 with Cbl inhibited this effect (Fig. 6, compare lanes 2, 3, and 4) (7, 27, 35). However, addition of ITSN1 reversed the inhibitory effect of Spry2 on Cbl, leading to enhanced ubiquitylation of endogenous EGFR (Fig. 6, compare lanes 4 and 5). These results demonstrate that ITSN1 overexpression disrupts Spry2-Cbl binding, resulting in enhanced Cbl activity for the activated EGFR.

Fig 6.

ITSN1 overexpression reverses the inhibitory effects of Spry2 on Cbl-mediated EGFR ubiquitylation. Overexpression of Cbl in COS cells results in enhanced EGFR ubiquitylation following EGF stimulation (compare lanes 2 and 3). Coexpression of Spry2 with Cbl reduces EGFR ubiquitylation even though Cbl levels are elevated even higher than in the absence of Spry2 overexpression (compare lanes 3 and 4). Coexpression of ITSN, however, reverses the effect of Spry2 resulting in increased EGFR ubiquitylation. The ratio of ubiquitylated EGFR to total EGFR was determined by densitometry and compared between samples. The results are shown in the graph below the Western blots. These results are representative of three independent experiments.

ITSN1 enhances the inhibitory effect of Spry2 P304A mutant.

ITSN1's SH3 domains bind the Pro-rich tail of Cbl (20) as well as Spry2 (Fig. 1 and 4). Since Cbl and Spry2 interact with each other, we next examined the effect of mutating the ITSN1-binding site of Spry2 on the interaction between Spry2 and Cbl. Spry2 P304A interacts with Cbl, as measured by BiFC (Fig. 7A, leftmost panels). Surprisingly, increasing ITSN1 expression resulted in increased interaction between Cbl and Spry2 P304A in a dose-dependent manner (Fig. 7A and B). The increase in the Spry2-Cbl BiFC signal was not due to changes in the expression of VN-Spry2 or VC-Cbl (Fig. 7C). To confirm these BiFC results, we again used epitope-tagged versions of these proteins and tested the effect of ITSN1 on coprecipitation of Spry2 P304A with Cbl. In agreement with the BiFC results, we observed that ITSN1 dose dependently increased association of Spry2 P304A with Cbl (Fig. 7D).

Fig 7.

ITSN1 binding to Cbl in the absence of Spry2 binding leads to enhanced Spry2-Cbl interaction and decreased EGFR ubiquitylation. (A) Interaction of Spry2 P304A with Cbl was measured by BiFC in the absence or presence of increasing ITSN1 levels as described in the legend to Fig. 5. ITSN1 overexpression results in enhanced binding of Spry2 P304A to Cbl. (B) Quantification of BiFC signal. Interaction of Spry2 P304A and Cbl was quantified as described previously (36). Results are the averages and SEM from three independent experiments. Asterisks indicate significant differences from the value for the VN-Spyr2 P304A+VC-Cbl sample (P < 0.05). (C) Western blotting demonstrates the expression of the various proteins. Both ITSN1 and Cbl are HA tagged. The differences in Spry2 P304A-Cbl interaction are not due to changes in the overall expression of these proteins. (D) Overexpression of HA epitope-tagged ITSN1 dose dependently enhances the binding of the Spry2 P304A mutant to Cbl. HA-Cbl was immunoprecipitated from cells, and the coprecipitation of Spry2 P304A was monitored by Western blot of Cbl precipitates. (Top) Western blots of anti-Cbl precipitates with the indicated antibodies. (Bottom) Western blots of cell lysates with the indicated antibodies.

Given this increased interaction between Spry2 P304A and Cbl in the presence of ITSN1, we next tested the consequence on EGFR ubiquitylation. Coexpression of Spry2 P304A with Cbl inhibited Cbl activity and thus decreased EGFR ubiquitylation (Fig. 8, compare lanes 3 and 4). These results are comparable to results with WT Spry2 (Fig. 6, compare lanes 3 and 4). However, in contrast to the results with wild-type Spry2, coexpression of ITSN1 with Spry2 P304A and Cbl further inhibited EGFR ubiquitylation compared to Spry P304A and Cbl, consistent with the increased interaction of Cbl and Spry2 P304A in the presence of ITSN1 (Fig. 8, compare lanes 4 and 5).

Fig 8.

ITSN1 overexpression enhances the inhibitory effects of Spry2 P304A on Cbl-mediated EGFR ubiquitylation. Overexpression of Cbl in COS cells results in enhanced EGFR ubiquitylation following EGF stimulation (compare lanes 2 and 3). Coexpression of Spry2 P304A with Cbl reduces EGFR ubiquitylation (compare lanes 3 and 4). Coexpression of ITSN1 enhanced the inhibitory effect of Spry2 P304A, resulting in a further decrease in EGFR ubiquitylation. The ratio of ubiquitylated EGFR to total EGFR was determined by densitometry and compared between samples. The results are shown in the graph below the Western blots. These results are representative of three independent experiments.

DISCUSSION

We have identified a novel molecular link between ITSN1 and Spry2 through two independent observations. First, a high-throughput Y2H screen for ITSN1 binding proteins identified multiple Spry2 clones as SH3-interacting proteins. Second, a peptide screen of SH3 domains from various proteins revealed ITSN1 (and ITSN2) as a potential interacting partner of Spry2. Our results demonstrate that Spry2, but not other Spry isoforms, is a bona fide ITSN1 target. Furthermore, this association is mediated predominantly through ITSN1's SH3 domains binding Spry2's C-terminal Pro-rich site (aa 304 to 310). Indeed, this Pro-rich sequence conforms to previously identified ITSN1 binding sites (3, 20, 37).

Our previous work demonstrated a novel role for ITSN1 in regulating Cbl-dependent ubiquitylation of the EGFR, resulting in increased degradation of the receptor following growth factor stimulation (20). However, the mechanism by which ITSN1 enhanced Cbl activity was unclear. The identification of Spry2 as an ITSN1 target provides a potential answer to this question. Cbl regulation is quite complex, involving posttranslational modifications as well as association of Cbl with numerous activators and inhibitors (28). Although ITSN1's ability to activate Cbl did not stem from alterations in Cbl binding to the EGFR, changes in Cbl stability, or altered tyrosine phosphorylation of Cbl, we proposed that ITSN1 activation of Cbl may occur through enhancing Cbl binding to an activator or inhibiting Cbl interaction with an inhibitor (20). Our current results demonstrate that ITSN1 regulates Cbl, in part, through disrupting the inhibitory effect of Spry2 on Cbl, thereby enhancing EGFR ubiquitylation by Cbl. The importance of this regulation by ITSN1 is highlighted by the finding that EGFR ubiquitylation is not necessary for internalization of the receptor but rather necessary for the sorting of the receptor in the multivesicular endosomes or bodies for degradation in the lysosome (6, 12). Thus, enhancing ubiquitylation of the EGFR leads to enhanced EGFR turnover, thereby altering EGFR signaling.

It should be noted that although the observed effects of ITSN1, Cbl, and Spry2 (Spry2 P304A) are rather modest, we likely underestimated the effects of these proteins on EGFR ubiquitylation, since we measured ubiquitylation of endogenous EGFR in the total population of cells yet were able to transfect only approximately 50% of cells. This approach allows us to measure the effects on endogenous receptors using endogenous ubiquitin and therefore avoids problems of uneven expression of epitope-tagged ubiquitin between samples (23). In addition, this approach also reduces the number of plasmids that are being transfected in any given sample, which also results in more consistent expression of the given proteins between experiments.

Our findings reveal a complex network of interactions between ITSN1 and the Pro-rich regions of both Cbl and Spry2 resulting in either activation or inhibition of Cbl, depending on how ITSN1 interacts with each of these components. Thus, modulating the interaction of ITSN1 with Spry2 and Cbl may lead to activation or repression of Cbl's ubiquitin ligase activity to regulate EGFR ubiquitylation. Both Cbl and Spry2 possess Pro-rich motifs that bind ITSN1 (Fig. 1D) (20). Surprisingly, disrupting the binding of ITSN1 to Spry2 enhanced interaction between Cbl and Spry2 P304A, leading to decreased EGFR ubiquitylation (Fig. 6). This result suggests that ITSN1 binding to the Pro-rich tail of Cbl may promote a conformational change that enhances the interaction of Spry2 with Cbl. While Spry2 binds Cbl through phosphotyrosine-dependent and -independent mechanisms (reviewed in reference 15), ITSN1 overexpression does not alter the tyrosine phosphorylation of Cbl following growth factor stimulation (20). Thus, we do not believe that the enhanced interaction of Spry2 P304A with Cbl is due to altered phosphorylation of Cbl. However, it is unclear whether ITSN1 overexpression alters the phosphorylation of Spry2 to facilitate interaction with Cbl.

ITSN1 regulates numerous biological processes, including endocytosis and cellular signaling (25). The modular structure of ITSN1 allows interaction with a variety of targets. Furthermore, intra- and intermolecular interaction of these domains appears to play an important role in ITSN1 function. For example, overexpression of ITSN1's SH3 domains inhibits the formation of clathrin-coated pits as well as ITSN1-regulated signaling pathways (29, 30, 32), indicating that SH3 domain availability must be strictly regulated to maintain proper ITSN1 function. Our data suggest that the EH and CC domains may negatively regulate SH3 domain availability as progressive NH₂-terminal deletions in ITSN1 enhanced binding to Spry2 as well as N-WASP. This regulation of SH3 binding may also have important implications for Cdc42 regulation by the long isoform of ITSN1 (ITSN1-L). ITSN1-L GEF activity is autoinhibited through an intramolecular interaction of the GEF domain with the linker region between the SH3E and the DH domain (16). Furthermore, interaction of ITSN1-L with N-WASP relieves this inhibition (13). Thus, EH binding to endocytic proteins, such as epsin (29), stonin (21), SCAMP1 (8), FCHo proteins (11), AP180 (34), and Dab (34), may enhance interaction of the SH3 domains with their targets to relieve this autoinhibition, thereby resulting in Cdc42 activation.

The activation of ITSN1 likely requires a complex of contributing proteins (25). While binding to targets as noted above may regulate ITSN1 function, localization also likely plays an important role in ITSN1 function. Interaction of ITSN1 with endocytic proteins facilitates ITSN1's translocation to the plasma membrane, where it participates in vesicle assembly (11). However, this recruitment may also allow cross talk with RTK-associated Cbl and regulation of receptor ubiquitylation. In addition, EH domain binding to components of the JNK MAPK pathway (1, 24) may also free the SH3 domains for interaction with various targets, such as Cbl and Spry2.

The identification of this novel ITSN1-Spry2 connection raises new questions in the pathophysiology of several diseases. ITSN1 has been implicated in the pathology of Down syndrome and Alzheimer's disease due to an increased expression of ITSN1 in patients and its participation in neuronal survival and differentiation (3, 5, 14, 39). There is a high comorbidity of the obstructive gastrointestinal disorder Hirschsprung disease in Down syndrome. Hirschsprung disease is caused by a failure of enteric nerve ganglia to migrate to the gut. Interestingly, Spry2 has been reported to regulate neurite outgrowth in the sympathetic neuron-like PC12 cells (10). Moreover, Spry2-deficient mice develop enteric nerve hyperplasia (31). The development of esophageal achalasia and intestinal pseudo-obstruction in these mice is reminiscent of Hirschsprung disease, and together these data suggest that pathogenesis of the disease may lie in the interaction between ITSN1 and Spry2.