NEDD4-1 is a proto-oncogenic ubiquitin ligase for PTEN

Abstract

The tumor suppressor PTEN, a critical regulator for multiple cellular processes, is mutated or deleted frequently in various human cancers. Subtle reductions in PTEN expression levels have profound impacts on carcinogenesis. Here we show that PTEN level is regulated by ubiquitin-mediated proteasomal degradation, and purified its ubiquitin ligase as HECT-domain protein NEDD4-1. In cells NEDD4-1 negatively regulates PTEN stability by catalyzing PTEN polyubiquitination. Consistent with the tumor-suppressive role of PTEN, overexpression of NEDD4-1 potentiated cellular transformation. Strikingly, in a mouse cancer model and multiple human cancer samples where the genetic background of PTEN was normal but its protein levels were low, NEDD4-1 was highly expressed, suggesting that aberrant upregulation of NEDD4-1 can posttranslationally suppress PTEN in cancers. Elimination of NEDD4-1 expression inhibited xenotransplanted tumor growth in a PTEN-dependent manner. Therefore, NEDD4-1 is a potential proto-oncogene that negatively regulates PTEN via ubiquitination, a paradigm analogous to that of Mdm2 and p53.

Figure 1. Polyubiquitination of PTEN in vivo and in vitro

(A) Polyubiquitination of transfected PTEN. C-terminal His-tagged PTEN was co-transfected with HA-tagged ubiquitin (HA-Ub) in 293T cells as indicated. The cells were treated with or without 25 μM proteasome inhibitor MG132 for 4 hours and then harvested and lysed. His-tagged PTEN was pulled down with Ni²⁺ beads, washed with 6 M guanidine as described in Experimental Procedures, and subjected to immunoblotting against HA-tag to detect ubiquitinated PTEN. (B) In vitro ubiquitination of PTEN. The in vitro assay was performed as described in Experimental Procedures, with individual components added as indicated. GST-PTEN was used as the substrate, and HeLa S-100 (HS100, 5 μl and 5 mg protein/ml) was required to provide the PTEN ubiquitin ligase (E3) activity.

Figure 2. Purification and identification of the PTEN ubiquitin ligase as NEDD4-1

(A) The purification scheme. The detailed procedure was described in Experimental Procedures. (B) The PTEN E3 activity in fractions (Frac, 2 μl each) from the final Mono Q column was measured by the in vitro PTEN ubiquitination assay. (C) The final Mono Q fractions (15 μl each) were resolved by SDS-PAGE and stained with silver. (D) The purified PTEN E3 activity can ubiquitinate HA-tagged recombinant PTEN. The purified E3 (2 μl) or HS100 (5 μl) was used in the reactions as indicated. The ubiquitinated PTEN was detected by immunoblotting against HA-tag. (E) The domain structure of NEDD4-1. Amino acid positions of individual domains are denoted by corresponding numbers. (F) Interaction of NEDD4-1 with PTEN in 293T cells determined by coimmunoprecipitation analysis. PTEN and HA-tagged NEDD4-1 (N4-HA) were transfected as indicated. After immunoprecipitation using anti-HA antibody, PTEN and N4 in the precipitates were detected by immunoblotting (the upper two panels). The cell lysate (before immunoprecipitation) was blotted for N4, PTEN, and tubulin (the lower three panels) as controls. (G) Direct physical interaction of PTEN with NEDD4-1 (N4) determined by a GST pull-down assay. GST-PTEN but not GST alone pulled down purified NEDD4-1.

Figure 3. NEDD4-1 but not NEDD4-2 is an ubiquitin ligase for PTEN

(A) NEDD4-1 but not NEDD4-2 was presented in the purified PTEN E3 fraction. Different volume of HeLa S-100 (HS100, 5 mg protein/ml) and purified PTEN E3 activity (E3) were subjected to immunoblot against NEDD4-1 and NEDD4-2 as indicated. (B) NEDD4-1 fractions but not NEDD4-2 fractions possess PTEN E3 activity. NEDD4-1 and NEDD4-2 in HS100 were separated by Mono Q chromatography as shown by immunoblotting (top panel). The PTEN E3 activity in 5 μl of individual fractions (fractions 13 and 14 as NEDD4-1 fractions; 19 and 20 as NEDD4-2 fractions; fraction 11 as a negative control) was measured (bottom panel).

Figure 4. Regulation of PTEN ubiquitination and degradation by NEDD4-1 in cells

(A) Overexpression of NEDD4-1 enhances polyubiquitination of PTEN. His-tagged PTEN, HA-Ub, and increasing amount of NEDD4-1 were co-transfected in 293T cells as indicated. The ubiquitination of His-tagged PTEN was detected as described in Figure 1A. Lower panel shows expression of NEDD4-1. (B) Overexpression of NEDD4-1 causes a decrease in PTEN protein level. As indicated, NEDD4-1 with a C-terminal HA-tag or vector alone was transfected in 293T cells by electroporation. After 24 hours, the cells were treated with or without 50 μM MG132 for 4 hours as indicated. Subsequently, the cells were harvested and lysed, and the endogenous PTEN of individual samples was detected by immunoblotting. β-Tubulin was used as the loading control. PTEN levels (relative to Tubulin) were quantitated by densitometry. (C) NEDD4-1 polyubiquitinates endogenous PTEN and targets it for proteasomal degradation. As described in Experimental Procedures, His-tagged ubiquitin and NEDD4-1-HA were transfected into 293T cells as indicated. The cells were treated with or without MG132 as indicated, and then harvested and lysed. Proteins modified with His-tagged ubiquitin were pulled down and subjected to immunoblotting against PTEN. (D) RNAi of NEDD4-1. The efficacy of several human NEDD4-1 RNAi constructs (iN4, A–C) was examined by immunoblotting against HA-tag in HeLa cells co-transfected with NEDD4-1-HA and indicated RNAi constructs. A LacZ RNAi construct (iLacZ) was used as control. (E) Elimination of NEDD4-1 by RNAi increases PTEN level in cells. Individual RNAi constructs were co-transfected into HeLa cells with PTEN-HA as indicated. The expressed PTEN-HA was detected by immunoblotting. (F) NEDD4-1 overexpression decreases PTEN stability. As detailed in Experimental Procedures, PTEN plasmid was cotransfected with NEDD4-1 or control plasmid as indicated, cells were treated with cycloheximide, and harvest at indicated time points. Subsequently, immunoblotting against PTEN, NEDD4-1 (both HA-tagged), and tubulin control was performed. (G) NEDD4-1 RNAi increases PTEN stability. PTEN was cotransfected with iN4A (NEDD4-1 RNAi) or control plasmid as indicated, cells were treated with cycloheximide, and harvest at indicated time points. Subsequently, immunoblotting against PTEN and tubulin control was performed.

Figure 5. NEDD4-1 potentiates cell transformation in a PTEN-dependent manner

(A) NEDD4-1 regulates AKT phosphorylation. RNAi plasmid iLacZ or iN4A was transfected into 293 cells by electroporation. Immunoblotting against NEDD4-1, PTEN, phospho-AKT (p-Akt), total AKT, and β-tubulin was performed as indicated. (B) Overexpression of NEDD4-1 in p53^−/− primary MEFs decreased endogenous PTEN level. The MEFs were transfected with vector control or NEDD4-1 (N4-HA) as indicated by electroporation. The cells were lysed 24 hours later. N4-HA and endogenous PTEN were detected by immunoblotting. β-tubulin was blotted as the loading control. (C) Overexpression of NEDD4-1 promotes Ras-induced cell transformation in p53^−/− primary MEFs. The p53^−/− primary MEFs were infected with retroviruses encoding for no ectopic protein (Vector), Ras, and/or NEDD4-1 (N4), as indicated. Soft-agar colony formation assays were performed as described in Experimental Procedures. The pictures showing the typical colony formation assay result. (D) Overexpression of NEDD4-1 promotes cell transformation in a PTEN-dependent manner. The colony formation assays were performed in both p53^−/− primary MEFs and p53^+/−Pten^−/− primary MEFs. The plot represents the results from three independent experiments with standard deviation.

Figure 6. Reverse correlation of NEDD4-1 and PTEN levels in cancer samples

(A) High levels of NEDD4-1 inversely correlate further decrease of PTEN levels in a mouse prostate cancer model. The panel shows typical result of PTEN, NEDD4-1 (N4), and p-Akt IHC staining of Pten^hy/− mouse prostate tissues. The IHC shows that high NEDD4-1 expression levels always correlate with low PTEN levels which in turn correspond with high p-Akt and prostate tumour initiation (P-value < 0.001 calculated by Chi-square test). Glands with tumor (tm) and normal glands (nl) are indicated. (B) The mRNA levels of NEDD4-1 and PTEN in a cohort of 87 invasive human bladder cancer samples. The mRNA levels were measured using Affymetrix 2.0 Plus Chip. (C) Reverse correlation of NEDD4-1 mRNA levels with PTEN protein levels in the human bladder cancer samples. The 87 samples were classified into two groups (low NEDD4-1, 40 samples, and high NEDD4-1, 47 samples) based on the NEDD4-1 mRNA level of each sample relative to the mean NEDD4-1 mRNA value of the whole cohort. Then the PTEN IHC status of the two groups was plotted. The inverse correlation of NEDD4-1 mRNA levels with PTEN protein levels in these samples is statistically significant with a P-value of 0.04 calculated by both Chi-square and Mann-Whitney tests.

Figure 7. NEDD4-1 RNAi inhibited xenotransplanted tumor growth in a PTEN-dependent manner

(A) The human prostate cancer cell lines, DU-145 (PTEN-positive) and PC3 (PTEN-negative), were stably infected with NEDD4-1 RNAi and the control LacZ RNAi retroviruses. The infected cells were subcutaneously injected into athymic nude mice and the xenotransplanted tumor growth was measured, as described in Experimental Procedures. (B) A Model for the potential function of NEDD4-1 in prostate tumorigenesis. After loss of one allel of PTEN, aberrant upregulation of NEDD4-1 function will further decrease PTEN protein level and thus accelerates cancer development. Upregulation of NEDD4-1 might lead to malignancy more rapidly than a complete loss of PTEN gene, because the latter will trigger INK4a and p53-dependent senescence (Chen et al., 2005). NEDD4-1 offers a new entry point for therapeutic intervention that could arrest tumorigenesis in cooperation with the senescence response.