Type I γ Phosphatidylinositol Phosphate 5-Kinase i5 Controls the Ubiquitination and Degradation of the Tumor Suppressor Mitogen-inducible Gene 6*

Mitogen-inducible gene 6 (Mig6) is a tumor suppressor, and the disruption of Mig6 expression is associated with cancer development. Mig6 directly interacts with epidermal growth factor receptor (EGFR) to suppress the activation and downstream signaling of EGFR. Therefore, loss of Mig6 enhances EGFR-mediated signaling and promotes EGFR-dependent carcinogenesis. The molecular mechanism modulating Mig6 expression in cancer remains unclear. Here we demonstrate that type I γ phosphatidylinositol phosphate 5-kinase i5 (PIPKIγi5), an enzyme producing phosphatidylinositol 4,5-bisphosphate (PtdIns(4,5)P2), stabilizes Mig6 expression. Knockdown of PIPKIγi5 leads to the loss of Mig6 expression, which dramatically enhances and prolongs EGFR-mediated cell signaling. Loss of PIPKIγi5 significantly promotes Mig6 protein degradation via proteasomes, but it does not affect the Mig6 mRNA level. PIPKIγi5 directly interacts with the E3 ubiquitin ligase neuronal precursor cell-expressed developmentally down-regulated 4-1 (NEDD4-1). The C-terminal domain of PIPKIγi5 and the WW1 and WW2 domains of NEDD4-1 are required for their interaction. The C2 domain of NEDD4-1 is required for its interaction with PtdIns(4,5)P2. By binding with NEDD4-1 and producing PtdIns(4,5)P2, PIPKIγi5 perturbs NEDD4-1-mediated Mig6 ubiquitination and the subsequent proteasomal degradation. Thus, loss of NEDD4-1 can rescue Mig6 expression in PIPKIγi5 knockdown cells. In this way, PIPKIγi5, NEDD4-1, and Mig6 form a novel molecular nexus that controls EGFR activation and downstream signaling.

EGF receptor (EGFR) 2 is a critical receptor tyrosine kinase that controls cell growth and differentiation during embryo-genesis and adult homeostasis (1). After stimulation by its agonists, EGFR mediates multiple downstream signaling cascades that control cell proliferation, apoptosis, migration, survival, angiogenesis, and other biological processes (2). The appropriate regulation of EGFR activation is critical to maintain normal physiology, and loss or overactivation of EGFR can lead to multiple diseases (2,3). The overexpression or aberrant kinase activity of EGFR could result in unregulated growth stimulation, tumorigenesis, and metastasis in various tumor types (4). Therefore, targeted therapies directed against EGFR have been a primary focus of antitumor drug development over the past 10 years (5). An understanding of the molecular mechanisms controlling EGFR activation is required to design efficient anti-EGFR therapies and to identify biomarkers predicting higher sensitivity to anti-EGFR drugs.
Mig6 (also known as RALT or Gene 33) is a widely expressed adaptor protein that directly binds to the kinase domain of EGFR to block its activation (6 -8). Agonist binding stimulates the formation of an asymmetric EGFR dimer by the intracellular kinase domains in which the carboxy-terminal lobe (C lobe) of one kinase domain induces an active conformation in the other (9). Mig6 inhibits EGFR activation by blocking formation of the activating EGFR dimer interface (6). Furthermore, Mig6 controls the endosomal trafficking of EGFR (10). Following agonist stimulation, activated EGFR is internalized, and subsequent trafficking events determine the fate of internalized EGFR, including recycling back to the plasma membrane or trafficking to the lysosomes for degradation (11,12). Mig6 ensures the sorting of internalized EGFR to late endosomes and the subsequent sorting to lysosomes for degradation (10). Mig6 can be transcriptionally induced by growth factors, cytokines, and cell stress (13)(14)(15)(16). Deletion of Mig6 leads to hyperactivated EGFR signaling and causes carcinogenesis in multiple organs (14,17). The disruption of Mig6 in mice accelerates initiation and progression of mutant EGFR-driven lung adenocarcinoma (18). Thus, Mig6 is a tumor suppressor, and its expression is down-regulated in various cancers (14). The molecular mechanisms controlling Mig6 expression in cancer remain to be clarified.
Type I ␥ phosphatidylinositol phosphate kinase (PIPKI␥) is an enzyme that generates PtdIns(4,5)P 2 (19). The PIPKI␥ gene is alternatively spliced, resulting in at least six protein variants expressed in humans, known as PIPKI␥i1-i6 (20). These splicing variants have the same N terminus and kinase domain, and they only differ by different extensions at the C terminus (20,21). These extensions allow for association with unique binding partners and allow for different PIPKI␥ splice variants to produce PtdIns(4,5)P 2 with distinct subcellular distributions necessary to perform specific biological functions. Distinct roles of different PIPKI␥ splice variants have been found in mediating or controlling EGFR signaling. For instance, PIPKI␥i2 specifically binds to talin at cell adhesions, and the production of PtdIns(4,5)P 2 modulates talin-integrin interaction to regulate inside-out integrin signaling and cell adhesion turnover (22). EGFR phosphorylates PIPKI␥i2 to control local PtdIns(4,5)P 2 levels at adhesions. In this way, PIPKI␥i2 is required for nascent adhesion formation at the leading edge to facilitate EGFRinduced cell directional migration (22). Unlike PIPKI␥i2, PIPKI␥i5 is localized at endosomes and is required for EGFR sorting from endosomes to lysosomes for degradation (23). Therefore, the loss of PIPKI␥i5 blocks the lysosomal trafficking and degradation of activated EGFR, which dramatically enhances and prolongs EGFR activation and downstream signaling such as AKT and ERK MAPK activation (23). These results indicate that PIPKI␥i5 is a critical regulator of EGFR.

Loss of PIPKI␥i5 Dramatically Decreases Mig6 Expression-
PIPKI␥i5 is required for down-regulating EGFR activation and signaling (23). However, the molecular mechanism by which PIPKI␥i5 controls EGFR is not fully understood. Mig6 is a suppressor of EGFR, and knockdown of PIPKI␥i5 by siRNA in MDA-MB-231 cells dramatically decreased Mig6 protein expression levels (Fig. 1, A and B). Consistently, loss of PIPKI␥i5 significantly enhanced and prolonged TGF␣-induced autophosphorylation of EGFR on tyrosine 1068 (Fig. 1, A and C), indicating increased levels of EGFR activation. Downstream AKT activation was also enhanced and prolonged by PIPKI␥i5 knockdown (Fig. 1, A and D). Knockdown of Mig6 by siRNA similarly promoted EGFR activation and AKT activation as the effects induced by PIPKI␥i5 knockdown (Fig. 1, A, C, and D).
These results indicate that PIPKI␥i5 can regulate EGFR signaling by modulating Mig6 expression.
To determine whether PIPKI␥i5 lipid kinase activity was required for modulation of Mig6 expression, a knockdown-rescue approach was developed. siRNA was used to knock down endogenous PIPKI␥i5, and then wild-type PIPKI␥i5 or a kinasedead mutant (PIPKI␥i5KD) construct containing siRNA-FIGURE 1. PIPKI␥I5 controls Mig6 expression and EGFR signaling. MDA-MB-231 cells were transfected with control siRNA, PIPKI␥i5 (I␥i5) siRNA, or Mig6 siRNA separately. 72 h after siRNAs transfection, cells were stimulated with TGF␣ (10 nM) for the times indicated. A, the EGFR protein level, EGFR activation, AKT activation, and Mig6 expression were detected. The following were quantified: Mig6 protein level in TGF␣-untreated cells (B), EGFR activation detected by Tyr(P)-1068 antibody (C), and AKT activation detected by Ser(P)-473 antibody (D). Quantification of the Mig6 protein level was normalized with the GAPDH level. Quantification of EGFR or AKT activation was normalized with the total EGFR or AKT level. GFP, HA-tagged wild-type PIPKI␥i5, or kinase dead-mutant PIPKI␥i5 was constructed into PWPT (Addgene) lentivirus vectors. GFP, HA-PIPKI␥i5 WT, or HA-PIPKI␥i5 KD was then expressed by lentivirus transfection in PIPKI␥i5 knockdown MDA-MB-231 cells, and the effects on Mig6 expression were detected by immunoblotting (E) and quantified (F). Error bars indicate mean Ϯ S.E. from three independent experiments. *, p Ͻ 0.05. resistant silent mutations was re-expressed using lentivirusmediated infection. Expression of wild-type PIPKI␥i5 but not PIPKI␥i5KD significantly rescued Mig6 expression in PIPKI␥i5 knockdown cells (Fig. 1, E and F). These results confirm the roles of PIPKI␥i5 in Mig6 expression and indicate that kinase activity with resulting production of PtdIns(4,5)P 2 is required for PIPKI␥i5 control of Mig6 expression.
To rule out possible PIPKI␥i5 siRNA off-target effects, another approach, clustered regularly interspaced short palindromic repeats (CRISPR)-cas9 genome editing (24,25), was used to disrupt PIPKI␥i5 expression. Two different sgRNA sequences were designed to mediate the genome editing of PIPKI␥i5 in a breast cancer cell line, MDA-MB-231, and a lung cancer cell line, H-1975. Single clones with CRISPR-cas9-mediated PIPKI␥i5 genome editing were selected from MDA-MB-231 and H-1975 cells. The single clone selected from MDA-MB-231 has an in-frame 28-bp deletion at the targeted PIPKI␥i5 locus in one allele, leading to full disruption, and an in-frame 9-bp deletion at another allele, leading to a deletion of three amino acids at the C terminus of PIPKI␥i5 ( Fig. 2A). Although it is not a full disruption of PIPKI␥i5, PIPKI␥i5 protein expression is dramatically decreased and Mig6 expression is significantly decreased in this clone, similar to the siRNA-induced PIPKI␥i5 knockdown cells (Fig. 2, B and C). The H-1975 single clone has an in-frame 8-bp and an in-frame 59-bp deletion at the PIPKI␥i5 locus in two different alleles separately (Fig. 2D). This is a full disruption of PIPKI␥i5, and the Mig6 expression in this clone is significantly decreased (Fig.   2, E and F). These results further confirm that PIPKI␥i5 controls Mig6 expression.
To demonstrate whether PIPKI␥i5 regulates steady-state Mig6 mRNA levels, a real-time PCR assay was performed to detect the mRNA levels of Mig6 in control or PIPKI␥i5 knockdown cells. As shown in Fig. 3A, knockdown of PIPKI␥i5 does not significantly change Mig6 mRNA levels. This suggests that PIPKI␥i5 may modulate Mig6 protein degradation but not the steady-state Mig6 mRNA levels. The efficiency of PIPKI␥i5 knockdown was confirmed by Western blotting (Fig. 3B). To define whether PIPKI␥i5-regulated Mig6 degradation is dependent on lysosomes or proteasomes, the lysosome inhibitor chloroquine or the proteasome inhibitor MG132 was used to treat the cells. Only inhibition of the proteasome function increased Mig6 expression in PIPKI␥i5 knockdown cells (Fig.  3C). This shows that PIPKI␥i5-regulated Mig6 degradation is dependent on proteasomes but not lysosomes. To confirm the efficiency of chloroquine in inhibiting lysosome function, the LC3A/BII level was detected because its degradation is dependent on lysosomes (26). To test the efficiency of MG132 in blocking proteasome function, the levels of SMAD ubiquitination regulatory factor 1 (Smurf1) and TGF␤ receptor I (TGF␤RI) were detected because their degradation is dependent on proteasomes (27,28). Chloroquine treatment increased the LC3A/BII level (Fig. 3D), and MG132 treatment enhanced Smurf1 and TGF␤RI levels (Fig. 3E). This validates the function of these two inhibitors. Exogenously expressed HA-tagged Mig6 was also degraded when endogenous PIPKI␥i5 was lost ( Fig. 3F). This further supports that PIPKI␥i5 controls Mig6 protein degradation but not the mRNA level.
To determine whether NEDD4-1 can use Mig6 as a substrate and mediate Mig6 ubiquitination, Mig6-NEDD4-1 interaction was first investigated. Endogenous Mig6 was immunoprecipitated from cell lysates and examined by Western blotting for association of endogenous NEDD4-1. As shown in Fig. 5A, NEDD4-1 was detected with the Mig6 complex. Their direct binding was confirmed using in vitro pulldown assays with purified GST-NEDD4-1 and His 6 -Mig6 (Fig. 5B). An in vivo ubiquitination assay was then performed to demonstrate whether NEDD4-1 can promote Mig6 ubiquitination. Myc-Mig6 and His 6 -ubiquitin were co-transfected into HEK-293 cells, and the ubiquitinated proteins were pulled down by Ni 2ϩ -NTA beads. Unubiquitinated proteins may also be pulled down by NTA beads via nonspecific binding or via interaction with ubiquitinated protein. To remove the unubiquitinated proteins, the NTA beads were washed under strong denaturing conditions as described under "Experimental Procedures." The amount of Mig6 in total ubiquitinated proteins was then detected by anti-Myc antibody via Western blotting. NEDD4-1 expression enhanced the Mig6 poly-ubiquitination level (Fig.  5C). To further confirm that NEDD4-1 directly ubiquitinates Mig6, an in vitro ubiquitination assay was used with purified NEDD4-1 and GST-Mig6 proteins. Addition of Mig6 and NEDD4-1 dramatically enhanced total ubiquitinated protein levels (Fig. 5D). Mig6 in the reaction system was immunoprecipitated by anti-Mig6 antibody, and then the levels of ubiquitinated Mig6 were detected by anti-ubiquitin antibody (Fig. 5E) via Western blotting. Addition of NEDD4-1 induced Mig6 ubiquitination (Fig. 5E). These results indicate that NEDD4-1 ubiquitinates Mig6.
PIPKI␥i5 Directly Interacts with NEDD4-1-The in vitro binding assay by purified His 6 -PIPKI␥i5 and GST-NEDD4-1 indicated the direct interaction of PIPKI␥i5 with NEDD4-1 (Fig. 6A). To determine whether the PIPKI␥i5-NEDD4-1 interaction is specific, the interaction of NEDD4-1 with other PIPKI␥ splicing variants was examined by co-IP assay. As shown in Fig. 6B, PIPKI␥i1 and PIPKI␥i2 do not have the ability to bind with NEDD4-1. This result demonstrated that the unique C terminus of PIPKI␥i5 is required for its association with NEDD4-1. Consistently, knockdown of PIPKI␥i2 does not affect Mig6 expression (Fig. 6C). This suggests that binding with NEDD4-1 is required for PIPKI␥i5 to modulate Mig6 expression. PIPKI␥i5KD still interacts with NEDD4-1 (Fig. 6B). This demonstrates that the kinase activity producing PtdIns(4,5)P 2 is not required for PIPKI␥i5-NEDD4-1 interaction. A series of truncations in the PIPKI␥i5-specific C terminus were made (Fig. 7A), and their ability to interact with NEDD4-1 was examined. The truncation mutant PIPKI␥i5 1-652 (the C terminus after amino acid 652 was deleted) lost the ability to interact with NEDD4-1 (Fig. 7B). This further supports that the C terminus of PIPKI␥i5 is required for NEDD4-1 interaction. NEDD4-1 is composed of three functional regions: an N-terminal C2 domain for membrane binding, a central region containing four WW domains for proteinprotein interactions, and a C-terminal homologous to the E6AP carboxyl terminus (HECT) domain for ubiquitin protein ligation (30). A series of NEDD4-1 truncation mutants were made (Fig. 7C), and the deletion of WW1 and WW2 decreased
Unlike the in vitro system, although kinase-dead mutant PIPKI␥i5 was expressed at a similar level as wild-type PIPKI␥i5, it was less efficient to decrease NEDD4-1-mediated Mig6 polyubiquitination compared with wild-type PIPKI␥i5 in the in vivo system (Fig. 8B). This indicates that kinase activity to produce PtdIns(4,5)P 2 is required for PIPKI␥i5 to efficiently modulate NEDD4-1-mediated Mig6 ubiquitination in vivo. This is consistent with the result that only wild-type PIPKI␥i5 but not the kinase-dead mutant can rescue Mig6 expression in PIPKI␥i5 knockdown cells (Fig. 1, E and F). Expression of wild-type PIPKI␥i5 but not the KD mutant decreased endogenous Mig6-NEDD4-1 interaction (Fig. 8, C and D). This suggests that PtdIns(4,5)P 2 produced by PIPKI␥i5 can disturb Mig6-NEDD4-1 interaction in vivo.
In the siRNA knockdown-rescue assay, expression of exogenous wild-type PIPKI␥i5 can rescue 70% but not full Mig6 expression in PIPKI␥i5 knockdown cells (Fig. 1, E and F). There might be an optimal PIPKI␥i5 expression level required to mediate its normal function. Inappropriate expression levels of a particular protein would either not rescue or could cause artificial effects (38). Although we used a lentivirus vector for attenuated PIPKI␥i5 expression compared with other types of protein expression vectors, the level of exogenous PIPKI␥i5 was still higher than the normal endogenous level. This may lead to some artificial effects that prevent exogenous PIPKI␥i5 to fully rescue endogenous PIPKI␥i5 function. Another possible reason is that individual cells may take up different amounts of siRNAversus cDNA-expressing constructs. In cells with a high siRNAversus cDNA ratio, the expression of PIPKI␥i5 may not be adequate to rescue Mig6 expression. Although it is not a full rescue, expression of exogenous wild-type PIPKI␥i5 does significantly  6 -Mig6 was mixed with 1 g of GST-NEDD4-1 or 1 g of GST and subjected to GST pulldown assay with glutathione-Sepharose 4B (GE Healthcare Life Sciences). C, different amounts of Myc-Mig6, combined with or without HA-NEDD4-1 and His 6 -ubiquitin, were transfected into HEK-293T cells, and then Mig6 ubiquitination was detected by an in vivo ubiquitination assay. D, the in vitro ubiquitination reaction of Mig6 was carried out in the presence of E1, UbcH5b, ubiquitin, and ATP with or without E3 NEDD4-1, and the whole protein ubiquitination level was tested by immunoblotting. E, Mig6 was purified by anti-Mig6 antibody from the in vitro ubiquitination reaction system and followed by Western blotting to detect the ubiquitination levels.
PIPKI␥i5 is required for EGFR sorting from the limiting membrane of endosomes into intraluminal vesicles (ILVs) of the multivesicular bodies, which is a key step for EGFR sorting to lysosomes for degradation (23). Hrs is a key component of the endosomal sorting complexes required for transport (ESCRT)-0 (49), which binds to ubiquitinated EGFR and recruits additional ESCRT components to mediate EGFR sorting into ILVs (50). PIPKI␥i5 facilitates Hrs interaction with Sorting Nexin 5 (SNX5), which blocks NEDD4-1-mediated Hrs ubiquitination and promotes Hrs association with EGFR to mediate EGFR sorting into ILV (23). Here we found that PIPKI␥i5 directly interacts with NEDD4-1 to modulate Mig6 ubiquitination and degradation. As such, the ubiquitination events are critical for PIPKI␥i5 in modulating EGFR. PIPKI␥i2 can be ubiquitinated by E3 ligase HECT domain E3 ubiquitin protein ligase 1 (HECTD1) in the N terminus at lysine 97 and resulted in PIPKI␥i2 degradation (51). Blocking PIPKI␥i2 ubiquitination by mutating lysine 97 to arginine perturbs focal adhesion assembly and cancer cell migration (51). PIPKI␥i5 has exactly the same N terminus and lysine 97 as PIPKI␥i2. Therefore, PIPKI␥i5 itself has the potential to be modified by ubiquitination, and this can modulate its function.
EGFR mutations have been identified in lung cancer, including the point mutation L858R and an in-frame short deletion mutation, del746 -750 (52)(53)(54)(55). These mutations render EGFR with constitutive kinase activity, and they are deficient for being sorted to the lysosomes for degradation (56,57). Mig6 still interacts with these EGFR mutants and plays essential roles in down-regulating signaling mediated by them in cancer. Loss of Mig6 dramatically accelerates the initiation and progression of EGFR L858R-driven lung adenocarcinomas (18). By modulating Mig6 expression, PIPKI␥i5 can regulate EGFR mutant-initiated cancer progression.
Mig6 not only interacts with EGFR but also interacts with other members of ErbB family receptors such as ErbB2 (7). Thus, Mig6 suppresses the signaling from ErbB2 (29). As an adaptor protein, Mig6 also interacts and modulates the function of many important components of different signaling pathways, including cdc42, IB␣, GRB2, Src, PI3K p85, and 14-3-3 (7, 58 -60). This suggest that PIPKI␥i5 can function in multiple signaling pathways by controlling the expression of Mig6.

Experimental Procedures
Cell Cultures and Transfection-The breast cancer cell line MDA-MB-231 was cultured using DMEM supplemented with 10% FBS. The lung cancer cell line H-1975 was cultured using 1640 medium supplemented with 10% FBS. For plasmid transfection, cells were transfected using Lipofectamine 2000 (Invitrogen) following the instructions of the manufacturer. For siRNA transfection, cells were transfected with Oligofectamine (Invitrogen) for 72 h following the instructions of the manufacturer.
CRISPR-cas9 Genome Editing-Genomic disruptions were created in MDA-MB-231 and H-1975 cells using the CRISPR)/ Cas9 system as described previously (61). The lentiCRISPR v2 plasmid (Addgene, plasmid no. 52961) was used to mediate the expression of the single guide RNA (sgRNA) and cas9 protein by lentivirus infection. The sgRNAs targeting the PIPKI␥i5specific sequence were designed using online software. Two different sgRNA-targeting sequences were chosen: 5Ј-TCA-CGGCCCGCAGTCGGCGA-3Ј and 5Ј-GCGGTCACTTA-CAGTCCCCG-3Ј. Lentivirus-infected cells were cultured in medium containing 1.0 g/ml puromycin for 2 days for selection. Surviving cells were then trypsinized, and a single cell was isolated by limited dilution of the selected cells in 96-well plates. Each single clone was passaged and expanded. The total DNA of each colony was extracted and genotyped. The genomic region surrounding the CRISPR/Cas9 target site for PIPKI␥i5 was PCR-amplified. The amplicons were cloned into the pCR TM 2.1 Vector (Invitrogen) and validated by sequencing.
RNA Extraction and Real-time PCR-MDA-MB-231 cells were harvested, and total RNA was extracted using the ISOLATE RNA mini kit (Bioline, catalog no. BIO-52072). 2 g of total RNA was used for cDNA reverse transcription with the kit (Applied Biosystems, catalog no. 4368814). SYBR Green master mix (Applied Biosystems, catalog no. A25742) was used for real-time PCR, and the results was analyzed by QUANT-STUDIO 3 software. The Mig6 mRNA abundance was normalized to the expression of GAPDH. Primers used for the PCR were as follows: 5Ј-CTACTGGAGCAGTCGCAGTG-3Ј (forward) and 5Ј-CCTCTTCATGTGGTCCCAAG-3Ј (reverse) for Mig6 and 5Ј-AATCCCATCACCATCTTCCA-3Ј (forward) and 5Ј-TGGACTCCACGACGTACTCA-3Ј (reverse) for GAPDH.
Immunoprecipitation and Immunoblotting-Immunoprecipitation was performed as described previously (62). Briefly, cells were harvested and lysed in 25 mM HEPES (pH 7.2), 150 mM NaCl, 0.5% Nonidet P-40, 1 mM MgCl 2 , and protease inhibitor mixture and then centrifuged and incubated with protein G-Sepharose and 2 g of antibody as indicated at 4°C for 4 h. The immunocomplexes were separated by SDS-PAGE and analyzed as indicated.
In Vitro E3 Ubiquitin Ligase Activity Assay-The E3 ubiquitin ligase activity assay was based on a method described in previous studies (64,65) with minor modifications. The ubiquitination reaction mixture contained 0.2 M E1, 1 M UBCH5C (E2), 0.5 M NEDD4-1, and 10 M ubiquitin in buffer (50 mM Tris-HCl (pH 7.5), 2 mM ATP, and 5 mM MgCl 2 ). After incubation at 30°C for 1 h, the reaction was terminated by SDS loading sample buffer. Mig6 poly-ubiquitination chains were prepared as above. To isolate Mig6 from the mixture, SDS was added to the reaction mixture to a final concentration of 1%, and the sample was incubated at 90°C for 10 min. After a 10-fold dilution of SDS with buffer (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Nonidet P-40, and 10% glycerol), Mig6 antibody (1.0 g, Santa Cruz Biotechnology) and protein G-Sepharose beads (5 l) were added. The mixture was incubated for 4 h at room temperature and spun at 1000 ϫ g to collect protein G-Sepharose beads. The beads were washed three times with buffer plus 0.1% SDS, and then proteins were eluted with SDS loading sample buffer and detected by immunoblotting with ubiquitin antibody or Mig6 antibody.
Protein-Lipid Overlay Assay-PIP strips (catalog no. P-6001, Echelon) are 2 ϫ 6 cm hydrophobic membranes that have been spotted with 100 picomoles of all eight phosphoinositides and seven other biological important lipids. PIP strips were first blocked with PBS including 3% BSA for 1 h and then incubated with GST-NEDD4-1 (WT) or GST-NEDD4-1 (Del C2) (1 g/ml) in PBS with 3% BSA. Subsequently, the PIP strips were washed extensively in PBS containing 0.1% Tween 20, and the bound protein was detected with anti-GST antibody by immunoblotting.
Statistics-All data analyses were performed using Sigma-Plot. Bar graphs represent mean Ϯ S.E. as indicated. Statistical significance was assessed using Student's t test.
Author Contributions-M. S. conducted most of the experiments and analyzed the results. J. C. conducted the CRISPR-cas9 genome editing and selected single clones. R. A. A. provided assistance with the preparation of the manuscript. Y. S. conceived the idea for the project, designed the experiments, and wrote the paper with M. S.