sipl1-facilitated pten ubiquitination contributes to its association with pten
TRANSCRIPT
1
2
3Q1
4Q25678
9
10111213
1415161718
34
3536
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
Cellular Signalling xxx (2014) xxx–xxx
CLS-08259; No of Pages 8
Contents lists available at ScienceDirect
Cellular Signalling
j ourna l homepage: www.e lsev ie r .com/ locate /ce l l s ig
SIPL1-facilitated PTEN ubiquitination contributes to its associationwith PTEN
OO
FJason De Melo a,b, Xiaozeng Lin a,b, Lizhi He a,b,c, Fengxiang Wei a,b,d, Pierre Major e, Damu Tang a,b,⁎a Division of Nephrology, Department of Medicine, McMaster University, Canadab Father Sean O'Sullivan Research Institute, St. Joseph's Hospital, Hamilton, Ontario, Canadac Massachusetts General Hospital (MGH), Harvard Medical School, Boston, MA 02114, USAd The Genetics Laboratory, Institute of Women and Children's Health, Longgang District, Shenzhen, Guangdong, PR Chinae Department of Oncology, McMaster University, Hamilton, Ontario, Canada
⁎ Corresponding author at: T3310, St. Joseph's Hospital,Ontario, L8N 4A6, Canada. Tel.: +1 905 522 1155x35168;
E-mail address: [email protected] (D. Tang).
http://dx.doi.org/10.1016/j.cellsig.2014.08.0130898-6568/© 2014 Published by Elsevier Inc.
Please cite this article as: J. De Melo, et al., C
Ra b s t r a c t
a r t i c l e i n f o19
20
21
22
23
24
25
26
27
Article history:Received 12 July 2014Accepted 17 August 2014Available online xxxx
Keywords:SIPL1/sharpinPTENUbiquitinationProtein–protein interaction
28
29
30
31
32
33
ECTED PPTEN is post-translationally modified by ubiquitin via association with multiple E3 ubiquitin ligases, includingNEDD4-1, XIAP, and WWP2. Despite the rapid progress made in researching the impact of ubiquitination onPTEN function, our understanding remains fragmented. Building on the previously observed interaction betweenSIPL1 and PTEN, we report here that SIPL1 promotes PTEN polyubiquitination via lysine 48 (K48)-independentpolyubiquitin chains. Substitution of the K48 residue of ubiquitin with arginine (R) enhanced SIPL1-mediatedPTEN polyubiquitination. In contrast, the K63R substitution significantly reduced it. The ubiquitin-like (UBL)domain is required for SIPL1-induced PTEN polyubiquitination. This post-translational modification promotedthe association of SIPL1 with PTEN. Elevated amounts of the SIPL1/PTEN complex were precipitated in 293Tcells co-transfected with PTEN, SIPL1, and ubiquitin compared to cells co-transfected with SIPL1 and PTENonly. Additionally, formation of the SIPL1/PTEN complex was inhibited when either lysine-less (K0) ubiquitinor K63R ubiquitin was co-transfected together with SIPL1 + PTEN. The PTEN component in the SIPL1/PTENcomplex contained polyubiquitin chains. The ubiquitination reaction may play a structural role, stabilizing theSIPL1/PTEN complex, as a ubiquitin binding-defective SIPL1 mutant (TFLV) is proficient in PTEN association.Collectively, we demonstrate that SIPL1 binds PTEN and enhances PTEN polyubiquitination which in turn pro-motes the interaction between SIPL1 and PTEN.
© 2014 Published by Elsevier Inc.
R
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
UNCO
R1. Introduction
In addition to dephosphorylating phosphotidylinositol-3, 4,5-triphosphate(PIP3) at the plasma membrane, the PTEN tumor sup-pressor also functions in the nucleus to regulate genome stability [1].The molecular basis enabling PTEN to play a role in multiple cellularprocesses is attributable to a complex network of regulation whichincludes ubiquitination. PTEN is ubiquitinated via its association withmultiple E3 ligases.
Among these E3 ligases, NEDD4-1 has been the most thoroughlyinvestigated. NEDD4-1 binds PTEN, causes PTEN polyubiquitination,and induces PTEN degradation [2]. In supporting the critical role ofPTEN in tumour suppression, NEDD4-1 displays oncogenic activityresulting from its induction of PTEN degradation [2]. NEDD4-1 expres-sion correlatedwith loss of PTEN in non-small-cell carcinoma [3]. Down-regulation of NEDD4-1 via SCFβ-TRCP-mediated degradation inhibitedtumorigenesis through enhancing PTEN function [4]. Deubiquitination
72
73
74
75
50 Charlton Ave East, Hamilton,fax: +1 905 521 6181.
ell. Signal. (2014), http://dx.d
of PTEN by USP13, a deubiquitinating enzyme, stabilized PTEN, andthereby inhibited breast cancer tumorigenesis [5]. Conversely, by pro-moting NEDD4-1-mediated PTEN polyubiquitination, a 34 kDa protein,p34SEI-1 exhibited oncogenic activities [6,7].
In addition to PTEN polyubiquitination, NEDD4-1 alsomodifies PTENvia monoubiquitination. This posttranslational modification stabilizesPTEN, and plays a role in PTEN nuclear translocation [8]. Throughactivation of NEDD4-1, Ndfip1 and Ndfip2 associate with PTEN, andpromote both polyubiquitination-mediated PTEN degradation [9]and monoubiquitination-facilitated PTEN nuclear trafficking [10].Nuclear PTEN contributes to genome stability [11] and cell cycle reg-ulation [12,13].
Further supporting the importance of PTEN ubiquitination in regu-lating its functions, its ubiquitination is also mediated by additional E3ligases, including XIAP [14], WWP2 [15], CHIP [16], ret finger protein(RFP) [17], and SHOP [18]. CHIP, in addition to ubiquitinating PTEN,also ubiquitinates activated AKT, targeting both for degradation [19]demonstrating the tight regulation of ubiquitination occurring withinthe PI3K pathway. Adding to this growing list of proteins that promotePTEN ubiquitination,we observed a role of SIPL1 in facilitating this reac-tion. SIPL1-induced PTEN polyubiquitination does not require K48 but
oi.org/10.1016/j.cellsig.2014.08.013
T
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168Q3
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
2 J. De Melo et al. / Cellular Signalling xxx (2014) xxx–xxx
UNCO
RREC
K63. The polyubiquitination does not affect PTEN protein stability butcontributes to the interaction between SIPL1 and PTEN.
2. Materials and methods
2.1. Cell lines and plasmids
HEK 293T and MCF7 cells were purchased from American TypeCulture Collection (ATCC; Manassas, VA), and cultured in DMEMmedia supplemented with 10% fetal bovine serum (FBS; SigmaAldrich; Oakville, ON) and 1% penicillin–streptomycin (Life Technolo-gies; Burlington, ON).
SIPL1 and its mutants (TFLV, ΔUBL, N- and C-fragment) were FLAGtagged and inserted into pcDNA 3 or pLHCX. PTEN was sub-clonedinto pcDNA 3. pRK5 HA-Ubiquitin (Addgene plasmid 17608), pRK5HA-Ubiquitin-K0 (missing all lysine residues; Addgene plasmid 17603),pRK5HA-Ubiquitin K48R (Addgene plasmid 17604) and pET3a UbiquitinK63R (Addgene plasmid 18898) were all obtained from Addgene(Cambridge MA), where they were deposited by Dr. Ted Dawsonand Dr. Cecile Pickart. Ubiquitin K63R was sub-cloned out of pET3aand into pRK5 and tagged with HA.
2.2. Calcium phosphate-based DNA transfection
HEK 293T cells were transfected with plasmids using a calciumphosphate transfection procedure. Briefly, 0.17 μg/cm2 (plate surface)of each plasmid was incubated in a solution containing 0.25 M CaCl2and HEPES buffered saline (2× HeBS; 0.28 M NaCl, 0.05 M HEPES,1.5 mM Na2PO4). Transfection was confirmed visually 24 and 48 hlater using GFP expression and fluorescence microscopy.
2.3. Retroviral infection
Empty vector and SIPL1 retrovirus were produced using a gag-pol (GP) and an envelope expressing vector (VSV-G) (Stratagene,Mississauga, ON). Briefly, the GP and VSV-G vectors were transientlyco-transfected with pLHCX or pLHCX SIPL1 into 293T cells using acalcium–phosphate transfection. The virus-containing medium washarvested 48 h later, filtered through a 0.45 μM filter, and centrifugedat 20,000 g for 120 min to concentrate the retrovirus. After treatmentwith the virus, MCF7 cells were selected for stable plasmid integrationwith hygromycin (0.5 mg/mL, Sigma Aldrich, Oakville, ON).
2.4. Cyclohexamide treatment
Cyclohexamide was purchased from Sigma Aldrich (Oakville, ON).Cells were treated with Cyclohexamide (100 μg/mL) for the indicatedtimes followed by protein lysate collection and western blot analysis.
2.5. Co-immunoprecipitation
Cells were harvested in a lysis buffer containing 20 mM Tris(pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100,25 mM sodium pyrophosphate, 1 mM NaF, 1 mM β-glycerophosphate,0.1 mM sodium orthovanadate, 1 mM PMSF, 2 μg/mL leupeptin and10 μg/mL aprotinin (Sigma Aldrich, Oakville, ON). 500 μg of proteinlysate was incubated with co-immunoprecipitation buffer (50 mMTris–Cl pH 7.5, 10 mM EGTA, 100 mM NaCl, 0.1% Triton X-100),30 μL of rProtein G Agarose Beads (Life Technologies; Burlington,ON) and 1 μg of antibody overnight at 4 °C. The following antibodieswere used; anti-FLAG M2 (Sigma Aldrich, Oakville, ON), anti-PTEN(A2B1) (Santa Cruz Biotechnology, Santa Cruz, CA) or controlMouse IgG. The agarose beads were washed eight times with co-immunoprecipitation buffer before boiling with a Laemmli samplebuffer to separate the proteins from the antibody. The sampleswere an-alyzed using western blot.
Please cite this article as: J. De Melo, et al., Cell. Signal. (2014), http://dx.d
ED P
RO
OF
2.6. Western blot analysis
50 μg of protein lysate was separated on SDS-PAGE gels and trans-ferred onto Amersham Hybond ECL nitrocellulose membranes(Amersham, Baied'Urfe, QC). Blots were blocked with 5% skim milkand incubated at 4 °C overnight with either anti-FLAG M2 (1:1000,Sigma); anti-FLAG (1:1000, Sigma); anti-PTEN (138G6) (1:1000,Cell Signalling, Danvers, MA); or anti-GAPDH (1:5000, Cell Signalling,Danvers, MA). The blots were then incubated with the correspondingHRP-conjugated secondary antibodies for 1 h at room temperature.Signals were detected using an ECL Western Blotting Kit (Amersham,Baied'Urfe, QC). Protein bands were quantified using ImageJ software(National Institutes of Health).
3. Results
3.1. SIPL1 promotes PTEN polyubiquitination involving lysine 63 (K63)
Wehave previously demonstrated an interaction between SIPL1 andPTEN, an association that requires the ubiquitin-like domain (UBL) ofSIPL1 [20]. SIPL1 (Sharpin) plays a role in mediating NEMO linearpolyubiquitination [21–23]. These observations suggested a possiblerole of SIPL1 in promoting PTEN ubiquitination. This is consistent withPTEN being modified by ubiquitination [1]. To examine if SIPL1 contrib-uted to PTEN ubiquitination, 293T cells were transiently transfectedwith an N-terminal HA-tagged ubiquitin, SIPL1, and PTEN (Fig. 1A). Aclear indication of PTEN polyubiquitination was observed in 293T cellsco-transfectedwith all three plasmids (Fig. 1A). PTEN polyubiquitinationwas confirmed by western blot examination of immunoprecipitatedPTEN with anti-PTEN (Fig. 1B, top panel) and anti-ubiquitin (Fig. 1B,bottom panel) antibodies.
Ubiquitin contains several lysine residues (K6, K11, K27, K29, K33,K48, and K63) and thus an array of polyubiquitin chains can be formedvia different lysine linkages in protein ubiquitination; among thesepolyubiquitin chains, those formed through K48 and K63 occur mostcommonly and have been thoroughly studied [24]. While the K48polyubiquitination promotes protein degradation via the proteasome,the K63-linked polyubiquitin chains mediates protein trafficking andprotein interactions [25,26]. Although SIPL1/Sharpin is an essentialcomponent in the LUBAC that adds linear polyubiquitin chains toNEMO [21–23], SIPL1-facilitated PTEN polyubiquitination was unlikelyto occur in this way as the ectopic ubiquitin used in our system wasN-terminal HA-tagged. We thus determined the contributions of K48and K63 in SIPL1-involved PTEN polyubiquitination. By using theN-terminal HA-tagged K48R, K63R, and K0 (all lysine residues mutatedto arginine residues) ubiquitin mutants [27,28], we were able to showthat while K48R ubiquitin enhanced PTEN polyubiquitination in com-parison to ubiquitin, both K63R and K0 ubiquitin mutants substantiallyreduced both PTENmonoubiquitination and polyubiquitination (Fig. 2).Taken together, the above observations reveal that K63 but not K48plays a major role in SIPL1-mediated PTEN polyubiquitination.
3.2. The ubiquitin-like (UBL) domain is required for SIPL1 to promote PTENpolyubiquitination
SIPL1 contains three domains, a pleckstrin homology-like super fold(PH-like SF), UBL, and Npl4 zinc finger (NZF) [20,29,30] (Fig. 3A). TheUBL domain is required for PTEN binding [20]. The NZF domain isrequired for SIPL1 binding to mono, linear and K63-linked ubiquitinchains. Substitution of the T358 and F359 residues with leucine(T358L) and valine (T359V) within the NZF domain produced theSIPL1-TFLV mutant (Fig. 3A); the mutant was shown to be defective inlinear polyubiquitination of NEMO, as the TFLV mutant was unable tobind ubiquitin [22]. To examine the effects of these structural elementson SIPL1's ability to induce PTEN polyubiquitination, the SIPL1-TFLV,SIPL1-ΔUBL (deletion of the UBL domain), SIPL1-N, and SIPL1-Cmutants
oi.org/10.1016/j.cellsig.2014.08.013
T
OF
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
Fig. 1. SIPL1 induces PTEN polyubiquitination. A) 293T cells were transiently transfected with PTEN, FLAG-tagged SIPL1, and HA-tagged ubiquitin (HA-Ub) in the indicated combinations(top) for 48 h, followed by western blot analysis for PTEN, SIPL1, and GAPDH using anti-PTEN, anti-FLAG (SIPL1), and anti-GAPDH antibodies. PTEN-Ub: polyubiquitinated PTEN. Exper-imentswere performed three times; typical images from a single repeat are included. B) 293T cells were co-transfectedwith PTEN, SIPL1 and ubiquitin, followed by cell lysate preparation,immunoprecipitation (IP) with control IgG (IgG) and anti-PTEN, andwestern blot examination using the indicated antibodies. The IP with IgG and anti-PTENwas performed side-by-sideand analyzed in the same gel. The bands were rearranged for presentation.
3J. De Melo et al. / Cellular Signalling xxx (2014) xxx–xxx
EC
were constructed (Fig. 3A). Consistent with our previous report thatSIPL1-ΔUBLwas PTEN-binding defective [20], removal of the UBL domainsignificantly reduced SIPL1's ability tomodify PTENvia polyubiquitinationin comparison with SIPL1 (Fig. 3B). The association with ubiquitin is alsocritical, as evidenced by the dramatically reduced ability of SIPL1-TFLV insupporting PTEN polyubiquitination (Fig. 3B). The inability of SIPL1-Nmutant, which did not contain a UBL (Fig. 3A), in inducing PTENpolyubiquitinationwas in linewith the requirement of UBL in PTEN bind-ing [20] and PTEN polyubiquitination (Fig. 3B). The results also suggestthat the PH-like SF motif may not be critical for PTEN polyubiquitination.While SIPL1-C contains the UBL (Fig. 3A), we were unable to assay itseffectiveness in the modification of PTEN as SIPL1-C was expressed at avery low level in our system, suggesting that the mutant is unstable.The same resultswere independently obtained by two individuals. Collec-tively, we provide evidence that SIPL1's association with both PTEN andubiquitin is required to facilitate PTEN polyubiquitination.
3.3. Polyubiquitination does not induce PTEN degradation
The most thoroughly investigated role of polyubiquitination is pro-tein degradation through the proteasome [31]. To examine the potential
UNCO
RR
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
SIPL1
GAPDH
PTENSIPL1
HA-Ub
+ + + + +-- + - - -
+ + + +
HA-K0 - - - -
HA-K48R - - -+ -HA-K63R - - -- +
+
PTEN
PTEN-poly Ub
PTEN-mono Ub
Fig. 2. SIPL1 facilitates PTEN polyubiquitination independently of the K48-linkage. 293T cellswere transiently transfected with PTEN, FLAG-tagged SIPL1, HA-tagged ubiquitin (HA-Ub),and the HA-tagged ubiquitin mutants for 48 h. Western blot was subsequently performed.Experiments were performed three times; typical images from a single repeat are shown.
Please cite this article as: J. De Melo, et al., Cell. Signal. (2014), http://dx.d
ED P
ROeffects of SIPL1-initiated polyubiquitination on PTEN stability, PTEN,
SIPL1, and ubiquitin were co-transfected into 293T cells in specific com-binations (Fig. 4A, the left three panels). The abundance of the PTENprotein was not apparently decreased in the presence of a well-knownprotein synthesis inhibitor, cycloheximide (CHX), with or without theco-transfection of either SIPL1 or SIPL1 with ubiquitin (Fig. 4A, the leftthree panels). Similar results were also obtained when PTEN was co-transfected with SIPL1 and either the K48R or K63R ubiquitin mutants(Fig. 4A, the right two panels).
To consolidate these observations, MCF7 cell lines were establishedwhich stably express an empty vector (−) or SIPL1 (Fig. 4B). CHX treat-ment did not visibly affect PTEN protein stability in either pLHCX(empty vector) or SIPL1 overexpressing MCF7 cells during a period of24 h (Fig. 4C). Taken together, these observations demonstrate thatSIPL1-caused PTEN polyubiquitination does not result in PTEN degrada-tion. These results are in accordance with a major role for K63 but notK48-linked chains in SIPL1-facilitated PTEN polyubiquitination (Fig. 2).
3.4. PTEN ubiquitination facilitates its interaction with SIPL1
It is understood that protein ubiquitination is involved in protein–protein interaction [26], and it has been demonstrated that SIPL1binds to PTEN [20]. These concepts together with the above observa-tions that SIPL1-induced PTEN polyubiquitination is unlikely affectingPTEN's stability (Fig. 4) indicate a potential involvement of PTENubiquitination in its association with SIPL1. To test this possibility,293T cells were transiently transfected with PTEN plus or minus SIPL1and ubiquitin; all transfections were at comparable efficiencies (Fig. 5A).Immunoprecipitation (IP) of PTEN confirmed PTEN ubiquitination onlyin cells transfected with PTEN, SIPL1, and ubiquitin (Fig. 5B, the anti-ubiquitin panel). The inability to detect SIPL1 in the anti-PTEN IP reaction(Fig. 5B, the SIPL1 panel, lane 7) was due in all likelihood to the adjacentSIPL1 (lane 8) signal being far too intense (Fig. 5B, the SIPL1 panel; seepanel C for details). Longer exposure of PTEN revealed clear PTEN bandspresent in the anti-SIPL1 IP reaction using lysates prepared from 293Tcells transfected with PTEN, SIPL1, and ubiquitin (Fig. 5B, the PTEN long-exposure panel, lane 8). A significantly reduced level PTEN could be co-IPed via SIPL1 using lysates prepared from cells transfected with PTENand SIPL1 (see the * marked band, lane 5, the PTEN long exposurepanel, Fig. 5B). To clearly examine the co-IP efficiency of PTEN throughSIPL1 and vice-versa, the co-IPs in panel B were re-analyzed side-by-side. A clear co-IP of SIPL1 via PTEN (Fig. 5C, up panel) and PTEN viaSIPL1 was demonstrated only in cells co-transfected with PTEN, SIPL1,and ubiquitin (Fig. 5C). Taken together, these observations revealan important role of PTEN polyubiquitination in the formation of thePTEN/SIPL1 complex.
oi.org/10.1016/j.cellsig.2014.08.013
RRECTED P
RO
OF
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
Fig. 3.Characterization of SIPL1-mediated PTENpolyubiquitination. A) Schematic presentations of SIPL1 and its SIPL1mutants constructed. PH-like SF; Pleckstrin homology-like super fold.UBL; Ubiquitin-like domain. NZF; Npl4 zinc finger. TFLV; T358L and F359V mutant. B) 293T cells were transiently transfected with PTEN, SIPL1, the indicated SIPL1 mutants, and HA-Ub,followed bywestern blot analysis for the indicated events. Experimentswere performed three times; typical images from a single repeat are presented. Note:wewere unable to detect theSIPL1-C mutant.
4 J. De Melo et al. / Cellular Signalling xxx (2014) xxx–xxx
UNCO
To further examine the contributions of ubiquitination to the associ-ation of PTEN and SIPL1, we co-transfected 293T cells with and withoutSIPL1 in the absence or presence of ubiquitin and its mutants (Fig. 6A).SIPL1was immunoprecipitated fromcells inwhich SIPL1was transfected(Fig. 6B, the SIPL1 panel). A long exposure revealed the presence ofpolyubiquitinated PTEN, which was co-IPed through SIPL1, in cellsco-transfected with either ubiquitin or K48R ubiquitin but not in cellsco-transfected with the K0 or K63R ubiquitin mutants (Fig. 6B, the mid-dle panel). It was clear that IP of SIPL1 precipitated polyubiquitinizedPTEN more efficiently than non-ubiquitinized PTEN (Fig. 6B, top panel,comparing PTEN in the Ub and K48R lanes to PTEN in other lanes).These results became more apparent after the normalization of co-IPedPTEN to IPed SIPL1 (Fig. 6C). Similar results were also obtained in thereverse co-IP, i.e. IP of polyubiquitinized PTEN precipitated moreSIPL1 compared to the co-IP of SIPL1 via IP of non-ubiquitinized PTEN(Fig. 6C). Interestingly, the presence of K0 and K63R reduced co-IP effi-ciency in comparison to cells transfected with PTEN + SIPL1 (Fig. 6C),suggesting that these ubiquitin mutants may interfere with endogenousubiquitin-mediated formation of the PTEN/SIPL1 complex. Collectively,
Please cite this article as: J. De Melo, et al., Cell. Signal. (2014), http://dx.d
we provide compelling evidence demonstrating an important role forpolyubiquitination in the formation of the PTEN/SIPL1 complex.
3.5. Polyubiquitination enhances but is not required for the formation of thePTEN/SIPL1 complex
Our previous demonstration of the inability of SIPL1-ΔUBL tobind to PTEN together with its incompetency in mediating PTENpolyubiquitination (Fig. 3) would suggest an essential role ofubiquitination in the interaction of PTEN with SIPL1. To test this pos-sibility, we took advantage that the SIPL1-TFLV mutant which wasincapable of causing PTEN polyubiquitination and examined its abil-ity to bind PTEN. At a comparable transfection efficiency (Fig. 7, leftpanel), the SIPL1-TFLV mutant associated with PTEN with compara-ble affinity to SIPL1, demonstrated by the seemingly equivalentamount of the SIPL1-TFLV/PTEN and SIPL1/PTEN complexes IPedvia either PTEN (Fig. 7, middle panel) or SIPL1 (or SIPL1-TFLV) (Fig. 7,right panel). As expected, neither SIPL1-ΔUBL nor PTEN-N boundPTEN (Fig. 7, middle and right panels). While PTEN-C was expressed
oi.org/10.1016/j.cellsig.2014.08.013
OO
F
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
Fig. 4. SIPL1-mediated PTEN polyubiquitination does not apparently induce PTEN degradation. A) 293T cells were transiently transfected as indicated for 24 h followed by treatment with100 μg/mL of CHX for the indicated period of time. Western blot analysis was then performed for the indicated proteins. B) MCF7 cells were stably infected with empty vector (−) andSIPL1 retrovirus. The expression of ectopic SIPL1 along with PTEN and GAPDH was examined by western blot. C) MCF7 pLHCX (empty vector) and SIPL1 cells were treated as indicated,followed by western blot examination for the indicated proteins.
5J. De Melo et al. / Cellular Signalling xxx (2014) xxx–xxx
at undetectable levels (Fig. 7, left, panel), IP of SIPL1-C co-precipitated alow level of PTEN (Fig. 7, right panel). These results further support oursuggestion that the SIPL1-C truncation mutant was unstable in oursystem. However, the presence of UBL domain and the intact NZFmotif, a region from which the TFLV mutant was generated, may facili-tate PTEN ubiquitination and thereby enabling the association betweenSIPL1-C and PTEN (seeDiscussion for details). Taken together, the aboveobservations support the notion that PTEN ubiquitination enhances itsbinding to SIPL1.
UNCO
RRECT
Fig. 5. Polyubiquitination promotes the binding of PTEN and SIPL1. 293T cells were transi48 h. Cell lysates were prepared and immunoprecipitation with anti-PTEN, anti-FLAG fA) Input; B) immunoprecipitation of the cell lysates in panel A with the indicated antibSIPL1 (IB/immunoblotting FLAG) and PTEN (IB PTEN) presented in panel B.
Please cite this article as: J. De Melo, et al., Cell. Signal. (2014), http://dx.d
D P
R4. Discussion
PTEN functions in multiple processes, in different cellular compart-ments, with and without the involvement of its PIP3 phosphatase activ-ity [1,32]. The nature of its multi-functional activity in such a diversifiedsetting is attributable to a complex network of regulation. Even withregard to ubiquitination, PTEN is associated with different E3 ubiquitinligases, including XIAP [14], WWP2 [15], SPOP [18], RFP [17], andNEDD4-1 [2] and its activator Ndfip1/2 [9]. These interactions resulted
Eently transfected with PTEN, PTEN + SIPL1, and PTEN + SIPL1 + ubiquitin (Ub) foror SIPL1, and control IgG (IgG) was performed, followed by western blot analysis.odies, followed by western blot analysis; and C) re-analysis of the co-precipitated
oi.org/10.1016/j.cellsig.2014.08.013
TD P
RO
OF
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
Fig. 6. Characterization of the ubiquitination-promoted interaction between SIPL1 and PTEN. A) 293T cells were transiently co-transfected with PTEN, SIPL1, ubiquitin, and its mutants asindicated. The comparable transfection rates were confirmed by western blot. B) The input in panel A was used for IP as indicated. The IPs were analyzed by western blot for PTEN andSIPL1. Short and long exposurewas used to show the amount and ubiquitination of PTEN that was precipitated through SIPL1 IP. C) Co-precipitated PTEN presented in the short exposurepanel was normalised against IP SIPL1, and graphed. In the same manner, co-precipitated SIPL1 was normalized to IPed PTEN and graphed.
6 J. De Melo et al. / Cellular Signalling xxx (2014) xxx–xxx
RREC
in PTEN being mono- and poly-ubiquitinated. To make this post-translational modification more complex, we demonstrate here a rolefor SIPL1 in facilitating PTEN ubiquitination.
Although SIPL1 is an essential component of the LUBAC complex,an E3 ligase, it is very unlikely that the SIPL1-mediated PTENpolyubiquitination observed in our system was in the linear form,as the N-terminal HA-tagged ubiquitin used does not support linearpolyubiquitination [23]. While our research does not exclude thepossibility that SIPL1 may cause PTEN modification via linearpolyubiquitination through the LUBAC complex, our results supportthe possibility that SIPL1 promotes non-linear PTEN polyubiquitination.As SIPL1may not possess E3 ligase activity, whether SIPL1 facilitates theactions of an existing PTEN E3 ubiquitin ligases or other E3 ubiquitin
UNCO
Fig. 7. Ubiquitination enhances but is not essential for the association of SIPL1 with PTEN. 293mutants. The input of individual cell lysate is shown (left panel). IP was performed, followed bywas undetectable, IP of SIPL1-C led to the co-precipitation of PTEN (right panel).
Please cite this article as: J. De Melo, et al., Cell. Signal. (2014), http://dx.d
E
ligase remain unknown. It will be intriguing to identify these E3 ubiqui-tin ligases. Their identification will shed light on signals regulatingSIPL1-facilitated PTEN polyubiquitination.
The influence of PTEN ubiquitination has been largely seen to affectthe stability of the PTEN protein and its nuclear trafficking, withpolyubiquitination destabilizing PTEN and monoubiquitination causingits nuclear transport [2,8]. Our research is unique, in that SIPL1-initiatedPTEN polyubiquitination does not cause PTEN degradation. Polyubiquitinchains can be formed through the K48, K63, and other K linkages; whileK48 linkage-based polyubiquitin chains commonly direct protein deg-radation via the proteasome [31], the K63-linked polyubiquitin chainsregulate signal transduction, transcription, DNA damage repair, andother functions. Both RIP1 and NEMO, in addition to being the target
T cells were transiently co-transfected with PTEN + ubiquitin together with SIPL1 or itswestern blot analysis as indicated (right two panels). Note: although the SIPL1-C mutant
oi.org/10.1016/j.cellsig.2014.08.013
T
OF
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387Q4
388
389
390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424
Fig. 8.A schematicmodel to illustrate a role of PTEN polyubiquitination in the formation of the PTEN/SIPL1 complex. SIPL1 and PTEN are able to form an unstable complex (A). By inducingPTEN polyubiquitination, SIPL1 associates with the modified PTEN with significantly elevated affinity (B).
7J. De Melo et al. / Cellular Signalling xxx (2014) xxx–xxx
UNCO
RREC
of linear ubiquitin chains, are also poly-ubiquitinated by K63-linked chains, activating their protein kinase activity, leading tothe activation of NFκB [22,23,33,34]. Myc is the target of K63-linked polyubiquitination, which results in increasing its transcrip-tion activity and promoting cell proliferation [35,36]. XIAP inducesMEKK2 activation via K63-linked polyubiquitination [37]. K63-polyubiquitination contributes to proliferating cell nuclear antigen(PCNA) and γH2AX-facilitated DNA lesion repair [38,39]. Theobservations that SIPL1 facilitates PTEN polyubiquitination via K48-independent linkages (Fig. 2) are thus consistent with SIPL1 expressionnot resulting in PTEN degradation via polyubiquitination. Our resultsare in line with a very recent report in which RFP induced K48-independent PTEN polyubiquitination and this modification did not af-fect PTEN stability and its trafficking [17].
However, the K48-independent polyubiquitination of PTEN inhibitedPTEN-derived PIP3 phosphatase activity [17]. This is consistent with thereported inhibition of PTEN phosphatase activity by ubiquitination(eithermono or poly) [40].We have previously observed that by bindingto PTEN, SIPL1 inhibited PTEN phosphatase activity [20]. Based on thisresearch, it is possible that SIPL1 may inhibit PTEN phosphatase activityvia inducing PTEN polyubiquitination. However, this does not excludethe possibility that association with PTEN alone may also negatively im-pact its phosphatase activity. It will be interesting to determine the con-tributions of SIPL1-mediated polyubiquitination, and its interaction withPTEN, to the reduction of PTEN function. The SIPL1-TFLV mutant is idealfor this task, as it binds PTEN without causing its polyubiquitination(Figs. 3, 7).
In addition to inhibiting PTEN phosphatase activity, SIPL1 may alsoreduce PTEN transportation into the nucleus. This possibility is support-ed by 1) that nuclear PTEN functions independently of its phosphataseactivity; 2) that SIPL1 is a predominant cytosolic protein [20]; and 3)that more importantly SIPL1-induced PTEN polyubiquitination dramat-ically enhanced the interaction between SIPL1 and PTEN (Figs. 5, 6).
One of the unique features of our research is the demonstration ofpolyubiquitination significantly enhancing the binding between SIPL1and PTEN. This enhancement was further supported by the reductionof the SIPL1/PTEN complex formed in the presence of the ectopic K0and particularly K63R ubiquitin (Fig. 6). These mutants likely reducethe complex formation through their interference of endogenousubiquitin-mediated formation of the SIPL1/PTEN complex. This presentsan intriguing question as to whether K63R ubiquitin may negativelyimpact SIPL1's activity in inhibiting PTEN function. As the SIPL1-TFLVmutant binds non-ubiquitinated PTEN with a comparable affinity asSIPL1 to the polyubiquitinated PTEN (Fig. 7), we prefer a model inwhich SIPL1 forms an unstable complex with PTEN (Fig. 8A); in the
Please cite this article as: J. De Melo, et al., Cell. Signal. (2014), http://dx.d
ED P
RO
presence of ubiquitin, SIPL1 induces PTEN polyubiquitination; and thismodification stabilizes the SIPL1/PTEN complex (Fig. 8B).
5. Conclusions
We demonstrate here that SIPL1 induces PTEN polyubiquitinationindependently of the K48 linkage. The process requires the UBL domainwhichmediates SIPL1's associationwith PTEN and SIPL1's ability to bindubiquitin via the NZF motif. The polyubiquitination does not affect thestability of the PTEN protein, but enhances its association with SIPL1.
Acknowledgments
Thisworkwas supported by a Canadian Institutes of Health Research(CIHR) grant (COP — 107971) to D. Tang.
References
[1] Y. Shi, B.E. Paluch, X. Wang, X. Jiang, J. Cell Sci. 125 (2012) 4687–4692, http://dx.doi.org/10.1242/jcs.093765.
[2] X. Wang, L.C. Trotman, T. Koppie, A. Alimonti, Z. Chen, Z. Gao, et al., Cell 128 (2007)129–139, http://dx.doi.org/10.1016/j.cell.2006.11.039.
[3] N. Amodio, M. Scrima, L. Palaia, A.N. Salman, A. Quintiero, R. Franco, et al., Am. J.Pathol. 177 (2010) 2622–2634, http://dx.doi.org/10.2353/ajpath.2010.091075.
[4] J. Liu, L.Wan, P. Liu, H. Inuzuka, J. Liu, Z.Wang, et al., Oncotarget 5 (2014) 1026–1037.[5] J. Zhang, P. Zhang, Y. Wei, H. Piao, W. Wang, S. Maddika, et al., Nat. Cell Biol. (2013),
http://dx.doi.org/10.1038/ncb2874.[6] S. Jung, C. Li, D. Jeong, S. Lee, J. Ohk, M. Park, et al., Int. J. Oncol. (2013), http://dx.doi.
org/10.3892/ijo.2013.2064.[7] S.-W. Hong, J.-H. Moon, J.-S. Kim, J.-S. Shin, K.-A. Jung, W.-K. Lee, et al., Cell Death
Differ. 21 (2014) 146–160, http://dx.doi.org/10.1038/cdd.2013.141.[8] L.C. Trotman, X. Wang, A. Alimonti, Z. Chen, J. Teruya-Feldstein, H. Yang, et al., Cell
128 (2007) 141–156, http://dx.doi.org/10.1016/j.cell.2006.11.040.[9] T. Mund, H.R.B. Pelham, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 11429–11434,
http://dx.doi.org/10.1073/pnas.0911714107.[10] Y. Li, L.-H. Low, U. Putz, C.-P. Goh, S.-S. Tan, J. Howitt, Traffic (2014), http://dx.doi.
org/10.1111/tra.12175.[11] W.H. Shen, A.S. Balajee, J. Wang, H. Wu, C. Eng, P.P. Pandolfi, et al., Cell 128 (2007)
157–170, http://dx.doi.org/10.1016/j.cell.2006.11.042.[12] B.H. Choi, M. Pagano, C. Huang, W. Dai, J. Biol. Chem. 289 (2014) 17951–17959,
http://dx.doi.org/10.1074/jbc.M114.559005.[13] M.S. Song, A. Carracedo, L. Salmena, S.J. Song, A. Egia, M. Malumbres, et al., Cell 144
(2011) 187–199, http://dx.doi.org/10.1016/j.cell.2010.12.020.[14] C. Van Themsche, V. Leblanc, S. Parent, E. Asselin, J. Biol. Chem. 284 (2009)
20462–20466, http://dx.doi.org/10.1074/jbc.C109.009522.[15] S. Maddika, S. Kavela, N. Rani, V.R. Palicharla, J.L. Pokorny, J.N. Sarkaria, et al., Nat.
Cell Biol. 13 (2011) 730–735, http://dx.doi.org/10.1038/ncb2240.[16] S.F. Ahmed, S. Deb, I. Paul, A. Chatterjee, T. Mandal, U. Chatterjee, et al., J. Biol. Chem.
287 (2012) 15996–16006, http://dx.doi.org/10.1074/jbc.M111.321083.[17] J.T. Lee, J. Shan, J. Zhong, M. Li, B. Zhou, A. Zhou, et al., Cell Res. 23 (2013) 552–564,
http://dx.doi.org/10.1038/cr.2013.27.[18] G. Li, W. Ci, S. Karmakar, K. Chen, R. Dhar, Z. Fan, et al., Cancer Cell 25 (2014)
455–468, http://dx.doi.org/10.1016/j.ccr.2014.02.007.
oi.org/10.1016/j.cellsig.2014.08.013
425426427428429430431432433434435436437438439440441442443444445446
447448449450451452453454455456457458459460461462463464465466467468
469
8 J. De Melo et al. / Cellular Signalling xxx (2014) xxx–xxx
[19] C.-H. Su, C.-Y. Wang, K.-H. Lan, C.-P. Li, Y. Chao, H.-C. Lin, et al., Cell. Signal. 23 (2011)1824–1830, http://dx.doi.org/10.1016/j.cellsig.2011.06.018.
[20] L. He, A. Ingram, A.P. Rybak, D. Tang, J. Clin. Invest. 120 (2010) 2094–2108, http://dx.doi.org/10.1172/JCI40778.2094.
[21] F. Tokunaga, T. Nakagawa, M. Nakahara, Y. Saeki, M. Taniguchi, S. Sakata, et al.,Nature 471 (2011) 633–636, http://dx.doi.org/10.1038/nature09815.
[22] F. Ikeda, Y.L. Deribe, S.S. Skånland, B. Stieglitz, C. Grabbe, M. Franz-Wachtel, et al.,Nature 471 (2011) 637–641, http://dx.doi.org/10.1038/nature09814.
[23] B. Gerlach, S.M. Cordier, A.C. Schmukle, C.H. Emmerich, E. Rieser, T.L. Haas, et al.,Nature 471 (2011) 591–596, http://dx.doi.org/10.1038/nature09816.
[24] D. Komander, M. Rape, Annu. Rev. Biochem. 81 (2012) 203–229, http://dx.doi.org/10.1146/annurev-biochem-060310-170328.
[25] L. Sun, Z.J. Chen, Curr. Opin. Cell Biol. 16 (2004) 119–126, http://dx.doi.org/10.1016/j.ceb.2004.02.005.
[26] Z.J. Chen, L.J. Sun, Mol. Cell 33 (2009) 275–286, http://dx.doi.org/10.1016/j.molcel.2009.01.014.
[27] S. Raasi, C.M. Pickart, Methods Mol. Biol. 301 (2005) 47–55, http://dx.doi.org/10.1385/1-59259-895-1:047.
[28] K.L. Lim, K.C.M. Chew, J.M.M. Tan, C. Wang, K.K.K. Chung, Y. Zhang, et al., J. Neurosci.25 (2005) 2002–2009, http://dx.doi.org/10.1523/JNEUROSCI.4474-04.2005.
[29] B. Stieglitz, L.F. Haire, I. Dikic, K. Rittinger, J. Biol. Chem. 287 (2012) 20823–20829,http://dx.doi.org/10.1074/jbc.M112.359547.
UNCO
RRECT
Please cite this article as: J. De Melo, et al., Cell. Signal. (2014), http://dx.d
F
[30] Y. Daigo, I. Takayama, S.M.Ward, K.M. Sanders,M.A. Fujino, J. Gastroenterol. Hepatol.18 (2003) 712–718, http://dx.doi.org/10.1046/j.1440-1746.2003.03046.x (pii).
[31] M.H. Glickman, A. Ciechanover, Physiol. Rev. 82 (2002) 373–428, http://dx.doi.org/10.1152/physrev.00027.2001.
[32] J.E. Fata, S. Debnath, E.C. Jenkins, M.V. Fournier, Int. J. Cell Biol. 2012 (2012) 379685,http://dx.doi.org/10.1155/2012/379685.
[33] K. Newton, M.L. Matsumoto, I.E. Wertz, D.S. Kirkpatrick, J.R. Lill, J. Tan, et al., Cell 134(2008) 668–678, http://dx.doi.org/10.1016/j.cell.2008.07.039.
[34] H. Zhou, I. Wertz, K. O'Rourke, M. Ultsch, S. Seshagiri, M. Eby, et al., Nature 427(2004) 167–171, http://dx.doi.org/10.1038/nature02273.
[35] S. Adhikary, F. Marinoni, A. Hock, E. Hulleman, N. Popov, R. Beier, et al., Cell 123(2005) 409–421, http://dx.doi.org/10.1016/j.cell.2005.08.016.
[36] I. Hammond-Martel, H. Yu, E.B. Affar, Cell. Signal. 24 (2012) 410–421, http://dx.doi.org/10.1016/j.cellsig.2011.10.009.
[37] G. Winsauer, U. Resch, R. Hofer-Warbinek, Y.M. Schichl, R. de Martin, Cell. Signal. 20(2008) 2107–2112, http://dx.doi.org/10.1016/j.cellsig.2008.08.004.
[38] R.K. Chiu, J. Brun, C. Ramaekers, J. Theys, L. Weng, P. Lambin, et al., PLoS Genet. 2(2006) e116, http://dx.doi.org/10.1371/journal.pgen.0020116.eor.
[39] E.J. Bennett, J.W. Harper, Nat. Struct. Mol. Biol. 15 (2008) 20–22, http://dx.doi.org/10.1038/nsmb0108-20.
[40] H. Maccario, N.M. Perera, A. Gray, C.P. Downes, N.R. Leslie, J. Biol. Chem. 285 (2010)12620–12628, http://dx.doi.org/10.1074/jbc.M109.072280.
ED P
RO
O
oi.org/10.1016/j.cellsig.2014.08.013