Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells

ARTICLE IN PRESS Cancer Letters ■■ (2016) ■■–■■

Contents lists available at ScienceDirect

Cancer Letters j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a n l e t

1 2 3 4 5 6

Q2 Original Articles

Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells Q1 Yingqiu Zhang a,b,1, Jinrui Zhang a,b,1, Congcong Liu b,c, Sha Du b, Lu Feng d, Xuelin Luan b,

Yayun Zhang b, Yulin Shi b, Taishu Wang b, Yue Wu b, Wei Cheng b, Songshu Meng b, Man Li a,*, Han Liu a,b,c,**

7 8

a

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

Department of Oncology, Second Affiliated Hospital, Institute of Cancer Stem Cell, Dalian Medical University, Dalian, China Institute of Cancer Stem Cell, Dalian Medical University, Dalian, China c Cancer Biotherapy & Translational Medicine Center of Liaoning Province, Dalian Medical University, Dalian, China d Department of Pathology, First Affiliated Hospital, Dalian Medical University, Dalian, China b

A R T I C L E

I N F O

Article history: Received 12 June 2016 Received in revised form 25 August 2016 Accepted 27 August 2016 Keywords: Neratinib ErbB2 HER2 Breast cancer Ubiquitylation Endocytosis

A B S T R A C T

Receptor tyrosine kinase ErbB2/HER2 is frequently observed to be overexpressed in human cancers, leading to over activation of downstream signaling modules. HER2 positive is a major type of breast cancer for which ErbB2 targeting is already proving to be an effective therapeutic strategy. Apart from antibodies against ErbB2, the small molecule tyrosine kinase inhibitor lapatinib has had successful clinical outcomes, and other inhibitors such as neratinib are currently undergoing clinical investigations. In this study we report the effects of lapatinib and neratinib on the mRNA and protein levels of the ErbB2 receptor. We provide evidence that neratinib-induced down regulation of ErbB2 occurs through ubiquitinmediated endocytic sorting and lysosomal degradation. At the mechanistic level, neratinib treatment leads to HSP90 release from ErbB2 and its subsequent ubiquitylation and endocytic degradation. Our findings provide novel insights into the mechanism of ErbB2 inhibition by neratinib. © 2016 Elsevier Ireland Ltd. All rights reserved.

40 Introduction

41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

ErbB2/HER2 belongs to the ErbB family of receptor tyrosine kinases, with amplification frequently observed in a number of malignancies [1]. ErbB2/HER2 amplification identifies a major subtype of breast cancer called HER2-positive breast cancer that accounts for approximately 25–30% of all breast cancer cases [2]. Targeting ErbB2 has proven to be an effective strategy in the clinical treatment of HER2-positive breast malignancies. Two approaches have been approved by the US Food and Drug Administration (FDA) for clinical use: humanized monoclonal antibodies targeting the extracellular domain of ErbB2 and small molecule tyrosine kinase inhibitors (TKI) that block the kinase activity of ErbB2 [3]. Antibody based therapies include trastuzumab, pertuzumab, and the antibody-drug conjugate ado-trastuzumab emtansine (T-DM1). Lapatinib is a tyrosine kinase inhibitor treatment and is reversible, targeting both ErbB2 and EGFR.

58 59 60 61 62 63 64

Q3

* Corresponding author. E-mail address: [email protected] (M. Li). ** Corresponding author. E-mail address: [email protected] (H. Liu). 1 These authors contributed equally.

Neratinib (HKI-272) is another small molecule inhibitor that blocks the kinase activities of EGFR and ErbB2 [4]. Unlike lapatinib, neratinib irreversibly binds to the kinase domain of EGFR and ErbB2 via covalent interaction with conserved cysteine residues: C773 of EGFR and C805 of ErbB2 [5,6]. Multiple clinical investigations of neratinib treatment of HER2-positive breast cancer and other solid tumors are ongoing, with promising results for future treatment of HER2-positive malignancies [2]. The cell surface levels of many receptor tyrosine kinases are tightly regulated by endocytosis, which internalizes membrane receptors into the cells and subsequently sorts them into lysosomes for degradation via the endosome and multivesicular body systems [7]. ErbB2 is normally endocytosis impaired and requires the chaperone HSP90 for stability on the cell surface [8,9]. HSP90 inhibitors have been reported to induce ErbB2 degradation through either proteasomal or lysosomal pathways, which are regulated by CHIPmediated post-translational modifications of ErbB2 by ubiquitin (ubiquitylation) [10–12]. The mechanistic studies of TKI lapatinib and neratinib mainly focus on the inhibition of receptor kinase activity and corresponding downstream signaling pathways, most prominently the RAS– MAPK and PI3K–AKT signaling cascades, both of which can be efficiently suppressed by lapatinib and neratinib. Conversely, the effects of lapatinib and neratinib on the expression levels of EGFR and ErbB2 receptors per se remain underexplored. In the present

http://dx.doi.org/10.1016/j.canlet.2016.08.026 0304-3835/© 2016 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

2

90 91 92 93 94 95

study, we report that treatment with lapatinib and neratinib leads to increased and decreased ErbB2 expression levels respectively, although ErbB2 mRNA is increased by both inhibitors. The two TKIs also induce the endocytosis of ErbB2. Mechanistically, neratinib triggers potent ubiquitylation and endocytic degradation of ErbB2 via HSP90 dissociation from ErbB2.

96

Materials and methods

97

Antibodies and other reagents

with BCA protein assays. One milligram of lysates was incubated with protein G-agarose and anti-ErbB2 antibody (Santa Cruz) for 4 hours at 4 °C. Beads were washed 3 times with YP-IP buffer (0.1% Nonidet P-40, 25 mM Tris-HCl pH 7.5, 150 mM NaCl), and proteins were eluted with 1.5X SDS PAGE sample buffer. Samples were analyzed by immunoblotting with anti-ErbB2 and anti-Ubiquitin antibodies. In coimmunoprecipitation assays, SKBR3 cells were treated as described above before lysis with NP40 lysis buffer (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.1% Nonidet P-40). Cleared lysates (1 mg per sample) were incubated with protein G-agarose and antiErbB2 antibody at 4 °C for 4 hours. Eluted protein samples were subjected to SDS PAGE analysis before immunoblotting assays with anti-ErbB2 and anti-HSP90 antibodies.

156 157 158 159 160 161 162 163 164 165 166 167

98 99 100 101 102 103 104 105 106 107 108 109 110 111 112

Mouse anti-ErbB2 (A-2), mouse anti-ErbB2 (9G6), mouse anti-LAMP1, rabbit antiEEA1, and rabbit anti-GAPDH (FL-335) antibodies were purchased from Santa Cruz Biotechnology (CA, USA). Mouse anti-GAPDH antibody was purchased from Proteintech (Wuhan, China). Goat anti-ErbB2 N-terminus antibody was purchased from R&D. Rabbit anti-HER2/ErbB2 (29D8), rabbit anti-phospho-S6RP, rabbit anti-phosphoAkt (Ser473), rabbit anti-phospho-MEK1/2 (Ser217/221), mouse anti-phospho-p44/ 42 MAPK (Erk1/2) (Thr202/Tyr204), rabbit anti-phospho-mTOR, rabbit anti-MEK1/ 2, rabbit anti-ERK1/2, rabbit anti-AKT, and rabbit anti-phospho-ErbB2 antibodies were obtained from Cell Signaling Technologies. Mouse anti-EEA1 antibody was purchased from BD Transduction. Mouse anti-Ubiquitin (P4G7) antibody was purchased from Covance. Mouse anti-Tubulin antibody was obtained from Sigma. Secondary goat anti-mouse and anti-rabbit, donkey anti-goat antibodies were obtained from LICOR. Cycloheximide was purchased from MP Biologicals. Bortezomib, neratinib (HKI272), lapatinib (GW-572016), GDC-0941, and trametinib were purchased from Selleck. Chloroquine, geldanamycin, and dynasore were obtained from Sigma.

113

Cell culture

114 115 116 117 118 119

Cell lines with ErbB2 overexpression were obtained from the American Type Culture Collection (ATCC) and maintained in a humidified incubator (Thermo) at 37 °C with 5% CO2. SKBR3 cells were cultured in McCoy’s 5A medium (Gibco, USA), while AU565 and HCC1954 were cultured in RPMI 1640 medium (Gibco, USA), all media were supplemented with 10% fetal bovine serum (Gibco, USA) and 1% penicillin/ streptomycin (Thermo Fisher Scientific).

120

Cell lysis and immunoblottings

121 122 123 124 125 126 127 128 129 130 131 132

Cultured cells were washed twice with ice-cold PBS before lysis using RIPA buffer (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 1% w/v Triton X-100, 0.1% w/v SDS, 1% sodium deoxycholate) supplemented with mammalian protease and phosphatase inhibitors (Sigma). Lysates were cleared with centrifugation and protein concentrations were determined by a BCA protein assay (Thermo). Equal amounts of protein were resolved by SDS-PAGE and transferred to nitrocellulose membranes (Merck Millipore, USA). Blots were incubated with 5% fat-free milk in PBS for an hour at room temperature and then incubated with primary antibodies at 4 °C overnight. Then, the blots were washed three times with PBS before incubation with LICOR 680 nm or 800 nm infrared secondary antibodies for 1 hour. Following the PBS washes, the blots were scanned on a LICOR Odyssey system. The acquired images were analyzed with Image Studio Version 4.0 according to manufacturer’s instructions.

133

Immunofluorescence and confocal microscopy

134 135 136 137 138 139 140 141 142 143

ErbB2 overexpressing cells were cultured on glass coverslips in 6-well plates. Following treatment, cells on coverslips were fixed with 3% paraformaldehyde, then permeabilized with 0.2% Triton X100, and blocked in 10% goat serum. The cells were incubated with primary antibodies before addition of fluorescent secondary antibodies (Invitrogen, USA) at room temperature. Finally, the coverslips were washed and mounted. Cell staining was examined using a fluorescent microscope (Leica, Germany). To examine protein subcellular co-localization, Alexa Fluor® 488- and 594-conjugated secondary antibodies (Invitrogen, USA) were used together in immunofluorescence assays. Confocal images were captured with a Leica laser scanning confocal microscope (TCS SP5).

144

Quantitative PCR

145 146 147 148 149 150 151

ErbB2-overexpressing breast cancer cell lines were treated with the indicated inhibitors before total RNA extraction using Trizol (Invitrogen, USA). cDNA was generated, and quantitative PCR was performed as previously described [13]. The qPCR primers used were as follows: ErbB2 (forward 5′-GAGTGTCAGCCCCAGAATG-3′ and reverse 5′-GTAGGAGAGGTCAGGTTTCAC-3′), and beta-actin (forward 5′CACCTTCTACAATGAGCTGCGTGTG-3′ and reverse 5′-ATAGCACAGCCTGGATAGC AACGTAC-3′).

152

Immunoprecipitation and co-immunoprecipitation assays

153 154 155

For endogenous immunoprecipitation experiments, SKBR3 cells were treated with 500 nM lapatinib or neratinib prior to lysis with RIPA lysis buffer supplemented with protease and phosphatase inhibitors. Protein concentration was determined

Flow cytometry

168

Cultured SKBR3 and AU565 cells were treated with 500 nM of lapatinib or neratinib (DMSO as control) before one million cells were harvested. For cell cycle analysis, the cells were washed with PBS and fixed using 70% ethanol, prior to staining with 50 μg/ml propidium iodide (PI). For apoptosis assays, the cells were processed for Annexin V and PI double staining using an apoptosis assay kit (KeyGEN Biotech, China) according to manufacturer instructions. The samples were analyzed using a flow cytometer (Accuri C6, BD Biosciences) and acquired data were analyzed with FlowJo version 7.6.1 (FlowJo, LLC, USA).

169 170 171 172 173 174 175 176 177

Statistical analysis

178

To determine significant differences, assays were carried out 3 times with independent biological repeats, and data were presented as the mean ± standard error of the mean (SEM). Significant differences were assessed via Student’s t test using GraphPad Prism Version 5.01; p < 0.05 was considered statistically significant.

179 180 181 182 183

Results Lapatinib and neratinib show opposite effects on ErbB2 levels Given that breast cancer with ErbB2/HER2 amplification is the primary target for lapatinib and neratinib, we used three ErbB2overexpressing breast cancer cell lines in this study: SKBR3, AU565, and HCC1954. We treated SKBR3 and AU565 cells with lapatinib and neratinib (both at 500 nM). As expected, the phosphorylation of ErbB2, mTOR, AKT, S6RP, MEK and ERK1/2 downstream of ErbB2 is potently inhibited, demonstrating the effectiveness of the inhibitors on ErbB2 kinase activity (Fig. 1A). However, ErbB2 expression levels were differently affected by lapatinib and neratinib, lapatinib increased ErbB2 abundance (1.72- and 1.63-fold in SKBR3 and AU565, respectively by 24 hours), whereas neratinib decreased ErbB2 expression (to 48.5% and 48.9% in SKBR3 and AU565, respectively), although both inhibitors showed similar effects on cell cycle distribution and apoptosis (Fig. 1B–D). We first wondered whether these inhibitors affect ErbB2 transcription, and we thus carried out quantitative PCR assays to measure ErbB2 mRNA levels following treatment. As described in Fig. 1E and F, ErbB2 mRNA levels increased following 6 hours of treatment with both lapatinib and neratinib. By 12 hours, lapatinib elevated ErbB2 mRNA to 249% and 216% relative to control samples in SKBR3 and AU565 cells, respectively, while neratinib led to increases of 159% (SKBR3) and 136% (AU565). It therefore seems that lapatinib-induced ErbB2 elevation can be attributed to enhanced transcription but that neratinibtriggered ErbB2 decreases cannot be explained at the mRNA level. Subsequently, we assessed the involvement of two major signaling cascades downstream of ErbB2, the RAS–MAPK and PI3K–AKT pathways, in lapatinib mediated regulation of ErbB2 expression. Using the MEK inhibitor trametinib and the PI3K inhibitor GDC0941, we specifically blocked the RAS–MAPK and PI3K–AKT cascades, respectively, and then compared ErbB2 levels to lapatinib treatment, which inhibits both pathways. In SKBR3 cells, both trametinib and GDC0941 treatment led to increases in ErbB2 expression, similar to lapatinib treatment (Fig. 1G); but in AU565 cells, GDC0914 treatment resulted in a partial increase of ErbB2 compared to lapatinib, while trametinib failed to alter ErbB2 levels (Fig. 1G).

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

B AU565

SKBR3

Lapatinib Neratinib

Lapatinib Neratinib 0 12 24

12

Relative ErbB2 (%)

A

24 hrs

0

12 24 12

24 hrs

IB:ErbB2

150KD

IB:pErbB2

150KD

IB:pmTOR

200KD 66KD

IB:AKT

45KD

IB:pMEK

45KD 45KD

IB:MEK IB:pERK1/2 IB:ERK1/2

35KD

IB:pS6RP

C

150 100

12 24 12 24 hrs Lapatinib Neratinib

*

AU565

*

150 100

** **

50 0

Control

1500

12 24 12 24 hrs Lapatinib Neratinib Neratinib

80 60 * * *

20

*

40

*

*

*

* 20

0

0

0

0

S G2/M SKBR3

G1

S G2/M AU565

G

150

100

50

50

0

0

*

150

100

6 12 hrs SKBR3

0

0

0

6 12 hrs AU565

IB:ErbB2

150KD 55KD

150KD 55KD 250 200 150 100 50 0

6 12 hrs AU565 AU565

C

200

Relative ErbB2 (%)

*

* ** *

C o La ntr pa ol G ti D n Tr C0 ib am 94 et 1 in ib

Relative ErbB2 (%)

Neratinib

*

0

6 12 hrs SKBR3

SKBR3

F 200

100

100

on t La rol pa t G inib D C Tr 094 am 1 et in ib

G1

*

200

**

200

on t La rol pa ti G nib D C Tr 094 am 1 et in ib

40

300

300

**

C

60

Lapatinib 400

Relative ErbB2 (%)

80

Control Lapatinib Neratinib

** **

Annexin V

E

300 200

IB:Tubulin

** *

100 0

C o La ntr pa ol G ti D n Tr C0 ib am 94 et 1 in ib

100

*

DNA content Neratinib

Relative ErbB2 (%)

*

DNA content Lapatinib

AU565

AU565

Counts 100

SKBR3 PI

1200

1200

DNA content Control

Percentage of cells

Lapatinib

SKBR3

Counts

1200

800

Q8

0

D 1000

223 224 225 226 227 228 229 230

** **

50

0.0

IB:GAPDH

35KD

SKBR3

*

*

200

200 Relative ErbB2 (%)

66KD

45KD

250

0.0

IB:pAKT

3

Fig. 1. Lapatinib and neratinib display opposite effects on ErbB2 expression. (A) SKBR3 and AU565 cells were treated with 500 nM lapatinib or neratinib for 12 and 24 hours before cell lysis and immunoblotting with the indicated antibodies. GAPDH is shown as a loading control. (B) The bar charts show quantification of ErbB2 intensities relative to control sample (0 hours) by percentage. (C) The cell cycle distribution of cells treated with lapatinib or neratinib at 500 nM for 24 hours, control cells were treated with DMSO. The column charts below show the quantitation of cells at each stage by percentage. (D) Control and the inhibitor treated cells (500 nM for 60 hours) were stained with Annexin V and PI to inspect apoptotic populations. (E, F) Cells were treated as above for the indicated times, and the RNA was extracted to compare relative ErbB2 mRNA levels (relative to actin) by quantitative PCR. (G) SKBR3 and AU565 cells were treated with lapatinib (500 nM), GDC-0941 (1 μM), trametinib (500 nM), or DMSO as a control for 24 hours before immunoblotting assays with ErbB2 antibody. Tubulin is shown as a loading control. The graphs below show the quantification of ErbB2 compared to the control. All error bars represent the standard error of the mean (n = 3), *p < 0.05; **p < 0.01.

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

4

A

SKBR3 Control CHX 0 12 24 0 12 24

HCC1954 Control CHX 0 12 24 0 12 24 Lapatinib

AU565 Control CHX 0 12 24 0 12 24

150KD

IB: ErbB2

55KD

IB: pAKT IB: GAPDH

300

100 0 0

B

6

12 18 24 hrs

300

200

200

100 0 0

SKBR3 Control CHX 0 12 24 0 12 24

6

100 0

12 18 24hrs

0

6

12 18 24 hrs

HCC1954 Control CHX 0 12 24 0 12 24 Neratinib

AU565 Control CHX 0 12 24 0 12 24

IB: ErbB2

55KD

IB: pAKT

35KD

IB: GAPDH

Relative ErbB2 (%)

150KD

120

120

120

80

80

80

40

40

40

0 0

6

12

18

0 24 hrs 0

6

12

18

0 24 hrs 0

Neratinib CHX+Neratinib

6

12

18

24 hrs

D

C

0h

SKBR3

SKBR3 AU565 HCC1954

12h

AU565

Lapatinib 6h

HCC1954

0h

231 232 233 234 235 236

Lapatinib CHX+Lapatinib p = 0.02

200

300 p = 0.01

p = 0.02

Relative ErbB2 (%)

35KD

Neratinib 6h

12h

Fig. 2. Lapatinib and neratinib treatment leads to endocytic degradation of ErbB2. SKBR3, AU565, and HCC1954 cells were pretreated with 50 μg/ml cycloheximide (CHX) to stop protein synthesis before the addition of lapatinib (A) and neratinib (B). At 12 and 24 hours, control and treated cells were lysed for immunoblotting with indicated antibodies. pAKT was probed to confirm inhibitory effects of lapatinib and neratinib (HCC1954 contains PI3KCA mutation); GAPDH was probed to confirm equal loading. The graphs below show quantifications of ErbB2 against treatment time. (C, D) Cells were treated as mentioned above for the indicated times and processed for immunofluorescence experiments with anti-ErbB2 antibody (green). Nuclei were stained with DAPI (blue). Examples of intracellular ErbB2 punctae are indicated with yellow triangles. Scale bar = 10 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

237 238 239 240 241 242 243 244 245 246 247

Lapatinib and neratinib induce endocytosis of ErbB2 To assess the impact of lapatinib and neratinib on the existing pool of ErbB2 protein, we eliminated the influence of newly synthesized ErbB2 with cycloheximide treatment and investigated the effects of the inhibitors on ErbB2 levels. Consistent observations from SKBR3, AU565, and HCC1954 cells showed opposite effects of lapatinib on ErbB2 expression in cells treated with cycloheximide compared to control: lapatinib downregulated ErbB2 expression in the presence of cycloheximide (Fig. 2A). The inhibition of protein synthesis with

cycloheximide did not significantly enhance neratinib-induced ErbB2 downregulation in SKBR3, AU565, and HCC1954 cells (Fig. 2B). We next carried out immunofluorescence experiments to examine ErbB2 localization following lapatinib and neratinib treatment. In untreated SKBR3 and AU565 cells, ErbB2 is predominantly localized in the cell membrane (Fig. 2C). After lapatinib exposure (6 and 12 hours), intracellular punctae of ErbB2 are visible, which indicate ErbB2 internalization (Fig. 2C). In untreated HCC1954 cells, ErbB2 is located on the cell surface and in the cytoplasm, while lapatinib treatment enhances cytoplasmic accumulation of ErbB2

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

248 249 250 251 252 253 254 255 256 257

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

A

D

SKBR3 ErbB2

EEA1

SKBR3 ErbB2

Merge

5

LAMP1

Merge

0h

0h

6h

6h

12h

12h

B ErbB2

C

EEA1

EEA1

AU565 ErbB2

Merge

LAMP1

Merge

0h

0h

6h

6h

12h

12h

F

HCC1954 ErbB2

258 259 260 261 262

E

AU565

Merge

HCC1954 ErbB2

LAMP1

Merge

0h

0h

6h

6h

12h

12h

Fig. 3. Internalized ErbB2 travels through early endosome to lysosome. SKBR3 (A), AU565 (B), and HCC1954 (C) cells were treated with neratinib (500 nM) for the indicated times and processed for confocal microscopy assays with anti-ErbB2 (green) and anti-EEA1 (red) antibodies. Representative confocal sections are shown with magnified insets to illustrate colocalization after treatment (both 6 and 12 hours). (D–F) Cells were treated as mentioned above and confocal microscopy was carried out with antiErbB2 and anti-LAMP1 antibodies. Representative confocal sections with insets show colocalization following neratinib exposure at 6 and 12 hours. Scale bar = 10 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

263 264 265 266 267 268 269 270 271 272 273 274

(Fig. 2C). Neratinib treatment has a similar but greater affect, and it also leads to ErbB2 endocytosis in SKBR3 and AU565 cells. This effect is observed especially in HCC1954 cells in which the membrane distribution of ErbB2 is markedly reduced following neratinib addition, which is accompanied by increased cytoplasmic formation of ErbB2 punctae (Fig. 2D). Internalized ErbB2 follows endosome-lysosomal pathway Having observed inhibitor-induced internalization of ErbB2, we then sought to investigate whether internalized ErbB2 follows the

classical endosome lysosomal pathway to degradation [14]. Using EEA1 as an early endosome marker, we carried out confocal microscopy experiments to examine the colocalization of endocytosed ErbB2 with early endosomes. We used neratinib for this experiment because it triggers more potent endocytosis of ErbB2. As shown in Fig. 3, in untreated SKBR3, AU565, and HCC1954 cells, ErbB2 mainly localizes to the cell surface (HCC1954 cells also show cytoplasmic staining), and no apparent colocalization with EEA1 is observed in SKBR3 and AU565 cells, though occasional colocalization is visible in HCC1954 cells. Following neratinib treatment (for 6 and 12 hours), ErbB2 is consistently endocytosed in all 3 ErbB2-overexpressing cell lines, and

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

275 276 277 278 279 280 281 282 283 284 285

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

6

A

B

SKBR3 LAMP1

Merge

C

AU565 ErbB2

LAMP1

Merge

HCC1954 ErbB2

LAMP1

Merge

CQ+Neratinib

CQ+Lapatinib

Control

ErbB2

286 287 288 289

6h

12h

6h

12h

Fig. 4. Inhibition of lysosome activity accumulates ErbB2 following lapatinib and neratinib treatment. SKBR3 (A), AU565 (B), and HCC1954 (C) cells were pretreated with 80 μM chloroquine for an hour prior to lapatinib and neratinib treatment (500 nM) for 6 and 12 hours. Cells were stained with ErbB2 (red) and LAMP1 (green) antibodies before examination by confocal microscope. Representative confocal micrographs from each cell line are shown, with insets illustrating colocalization. Scale bar = 10 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325

the colocalization of internalized ErbB2 with EEA1 is evident in SKBR3 and AU565 and is further enhanced in HCC1954 cells (Fig. 3A–C). We then investigated whether these internalized ErbB2 proteins were sorted to lysosomes for degradation. We used LAMP1 as a marker for lysosomes and performed confocal microscopy experiments. As expected, colocalization of internalized ErbB2 with LAMP1 was observed following neratinib treatment in all three ErbB2-overexpressing cell lines (Fig. 3D–F). To further confirm that these inhibitors induced lysosomal degradation of ErbB2, we applied chloroquine to block lysosomal function and then treated the breast cancer cells with lapatinib or neratinib to induce ErbB2 endocytosis. As illustrated in Fig. 4, internalized ErbB2 accumulates in aberrantly enlarged lysosomes following lapatinib or neratinib treatment (especially in SKBR3 and AU565 cells where large vacuoles are visible), as can be seen by apparent colocalization with lysosomal marker LAMP1 (Fig. 4A–C).

detected a strong ubiquitin signal in ErbB2 immunoprecipitates from SKBR3 cells treated with neratinib at both 6 and 12 hours but not in the control or lapatinib-treated samples. This observation provides a compelling explanation for why neratinib induced more potent endocytosis and degradation of ErbB2 and also suggests the mechanism through which neratinib downregulates ErbB2. Since the 26S proteasome subunit POH1 regulates the ubiquitylation status of ErbB2, and since the blockage of proteasome activity increases ErbB2 ubiquitylation, we wondered whether proteasome activity is also involved in the regulation of neratinib-induced ErbB2 ubiquitylation [15]. To test this hypothesis, we pretreated SKBR3, AU565, and HCC1954 cells with the proteasome inhibitor PS-341 (Velcade) prior to neratinib treatment and compared the down regulation of ErbB2. The results from the quantitation of our immunoblots show that PS-341 stimulates ErbB2 degradation, providing support for the predicted deubiquitylation role of proteasome in this process (Fig. 6B–D).

Neratinib induced ErbB2 endocytosis is dynamin independent but regulated by ubiquitylation

343 Neratinib dissociates HSP90 from ErbB2 to trigger its ubiquitylation

Since GTPase dynamin often functions in receptor internalization by pinching off vesicles from cell membranes, we evaluated whether it was implicated in neratinib-induced ErbB2 downregulation. We treated SKBR3 and AU565 cells with the dynamin inhibitor dynasore before the addition of neratinib. The results from our immunoblotting experiments show that ErbB2 downregulation in dynasore-pretreated cells is not significantly affected compared to the control group (Fig. 5A and B). Consistently, in immunofluorescence assays ErbB2 internalization is still evident in SKBR3 and AU565 cells treated with dynasore (Fig. 5C and D). These results collectively suggest that neratinib-induced ErbB2 endocytosis is dynamin independent. Because ubiquitin plays pivotal roles during receptor endocytosis, we examined the ubiquitylation status of ErbB2 following lapatinib or neratinib exposure [14]. As shown in Fig. 6A, we

326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342

It is well established that the HSP90 chaperone protein maintains ErbB2 stability at the cell surface via constant binding and that dissociation of HSP90 recruits HSP70 and CHIP to ErbB2 to cause its ubiquitylation [16]. This prompted us to speculate that neratinib might cause ErbB2 ubiquitylation via the release of HSP90. We first tested whether treatment with an HSP90 inhibitor (geldanamycin) and neratinib had an additive effect on ErbB2 degradation. When SKBR3 and AU565 cells were treated with geldanamycin, neratinib, or both, only a very limited additive effect of neratinib and geldanamycin was observed (Fig. 7A and B). More importantly, we set up co-immunoprecipitation experiments to examine the amounts of HSP90 associated with ErbB2 under various treatment conditions. As illustrated in Fig. 7C, similar amounts of HSP90 (91%) were pulled down with ErbB2 from SKBR3 cells treated with lapatinib

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

B

SKBR3 + Neratinib Control Dynasore 0 6 12 0 6 12 h

A

AU565 + Neratinib Control Dynasore 0

IB: ErbB2

150KD

7

6

12

0

6

12 h IB: ErbB2

150KD

55KD

IB: pAKT

55KD

IB: pAKT

35KD

IB: GAPDH

35KD

IB: GAPDH

DynasoreDynasore+

80 40 0 0

6

120 Relative ErbB2 (%)

Relative ErbB2 (%)

120

80 40 0 0

12 h

C

DynasoreDynasore+

6

12 h

SKBR3 ErbB2

Neratinib DAPI

Dynasore+Neratinib Merge

ErbB2

DAPI

Merge

0h

6h

12h

AU565

D ErbB2

Neratinib DAPI

Dynasore+Neratinib Merge

ErbB2

DAPI

Merge

0h

6h

12h

360 361 362 363 364

Fig. 5. Neratinib induced ErbB2 endocytosis is dynamin activity independent. (A, B) SKBR3 and AU565 cells were pretreated with the dynamin inhibitor dynasore (100 μM, 30 minutes) or DMSO as control before neratinib exposure for 6 and 12 hours. Cell lysates were analyzed by immunoblotting with the indicated antibodies. The pAKT signal shows the activity of neratinib and GAPDH shows equal loading. (C, D) SKBR3 and AU565 cells were treated as described above. During immunofluorescence assays, cells were stained with ErbB2 antibody (green) and DAPI (blue) to show the nucleus. Examples of intracellular ErbB2 punctae are shown by yellow arrows. Scale bar = 10 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

365 366 367 368 369

compared to untreated cells, while remarkably reduced amounts of HSP90 (57.2%) co-immunoprecipitated with ErbB2 in cells treated with neratinib, suggesting weakened interaction of ErbB2 with HSP90 under neratinib treatment.

Discussion Breast cancer remains the most frequently diagnosed malignancy in women, accounting for approximately 30% of estimated new

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

370 371 372 373

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

8

A

B

6

ib

SKBR3

in

Velcade+

at

Velcade-

er N

C

La

on

pa

tro

tin

l

ib

Input

12

6

Neratinib 0

12 hrs IB: ErbB2

150KD 55KD

IB: Tubulin

6

12 0

6

hrs

12

IB:ErbB2

150KD 55KD

IB:Tubulin

12

6

12 hrs IB: Ub

150KD 116KD

6

12 0

150KD 55KD

Relative ErbB2 (%)

AU565

374 375 376 377 378

6

12 hrs

55KD

12 hrs

12 hrs

6

Neratinib 0

IB:Tubulin

40

6

0

150KD

80

0

0

IB:ErbB2

VelcadeVelcade+

120

40

HCC1954 Velcade+ Velcade6

12 0

6

12 hrs IB:ErbB2 IB:Tubulin

HCC1954 Relative ErbB2 (%)

Neratinib 0

VelcadeVelcade+

80

D

AU565 Velcade+ Velcade-

0

120

IB: ErbB2

150KD C

SKBR3

ib in at er

Relative ErbB2 (%)

6

N

C

La

on

pa

tro

tin

l

ib

ErbB2 IP

VelcadeVelcade+

120 80 40 0

0

6

12 hrs

Fig. 6. Neratinib triggers potent ubiquitylation of ErbB2 that is regulated by the proteasome. (A) SKBR3 cells were treated with 500 nM of lapatinib or neratinib for 6 and 12 hours before lysis. ErbB2 was immunoprecipitated from the lysates and samples were analyzed by immunoblotting with anti-ubiquitin and anti-ErbB2 antibodies. Ubiquitin smears are observed from neratinib treated samples. Tubulin is used as loading control. (B–D) SKBR3, AU565, and HCC1954 cells were pretreated with proteasome inhibitor PS-341 (Velcade) at 500 nM for 30 minutes or DMSO as control prior to neratinib addition (6 and 12 hours). Cell lysates were analyzed by immunoblotting with ErbB2 antibody. Tubulin shows equal loading. The graphs below show quantification of relative ErbB2 intensities compared to 0 hours.

379 380 female cancer cases in the US [17]. Breast cancer can be divided into 381 four subtypes: luminal A, luminal B, basal like, and HER2 positive. 382 ErbB2/HER2 overexpression confers results in poor prognosis for this 383 HER2-positive subtype, while offering a druggable therapeutic target. 384 In this regard, the FDA approved ErbB2 targeting therapies, includ385 ing trastuzumab (Herceptin), a monoclonal antibody and lapatinib 386 (Tykerb), a small molecule inhibitor, have made great progress in 387 increasing patient survival. Nonetheless, resistance develops fol388 lowing treatment in many cases, and hence, there is an urgent need 389 for alternative therapeutic strategies to prolong patient life 390 expectancy. 391 In the present study, we investigated the influence of the small 392 molecule ErbB2 inhibitors lapatinib and neratinib on the levels of 393 ErbB2 itself. Interestingly, we observed the opposite effects of these 394 inhibitors on ErbB2 expression, although with both elevated mRNA 395 Q4 levels of ErbB2. It is noteworthy that lapatinib treatment results in 396 a greater increase in ErbB2 mRNA expression compared to neratinib. 397 When we used trametinib and GDC-0941 to block the RAS–MAPK 398 and PI3K–AKT pathways separately, both inhibitors exhibited similar

effects on ErbB2 levels compared to lapatinib in SKBR3 cells but not in AU565 cells, where only GDC-0941 shows partial influence, indicating that the feedback regulation of ErbB2 expression is differentially regulated within various cellular environments. When protein synthesis is blocked, lapatinib treatment also leads to downregulation of ErbB2 levels, although to a much lesser extent than neratinib. Thus, both lapatinib and neratinib display two disparate effects toward ErbB2 expression, increases in ErbB2 mRNA compensate for the endocytic loss of ErbB2 in the case of lapatinib, and endocytic degradation of ErbB2 overwhelming the ErbB2 mRNA elevation in the case of neratinib, leading to opposite overall effects Q5 on ErbB2 levels. With a series of immunofluorescence and confocal microscopy experiments, we confirmed that lapatinib and neratinib induced the internalization and lysosomal degradation of the ErbB2 receptor. Inhibition of lysosomal degradation with chloroquine accumulates internalized ErbB2 in huge lysosome vacuoles. Using immunoprecipitation assays, we determined that neratinib induced more potent ErbB2 endocytosis due to strong ubiquitylation of ErbB2 following

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417

ARTICLE IN PRESS Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

SKBR3

A Neratinib

GA

GA+Neratinib

0 3 6 9 0 3 6 9 0 3 6 9 hrs IB:ErbB2 150KD IB:Tubulin 55KD Relative ErbB2 (%)

120

Neratinib GA GA+Neratinib

80

P

ErbB2

Lapatinib

P

P

P

Neratinib

Ub Ub Ub Ub

Endocytosis

Endocytosis

40 0

0

3

6

9 hrs

PI3K

MEK

AKT

ERK lysosomal degradation

lysosomal degradation

AU565

120

Neratinib GA GA+Neratinib

80

0

C

Fig. 8. A working model depicting lapatinib and neratinib regulation of ErbB2 expression. Both lapatinib and neratinib have dual roles in regulating ErbB2 levels. Through blocking downstream PI3K–AKT and RAS–RAF–MEK–ERK pathways, lapatinib and neratinib elevate ErbB2 transcription (lapatinib induces more than neratinib). Conversely, lapatinib and neratinib treatment leads to ErbB2 endocytosis and subsequent lysosomal degradation. Neratinib triggers potent ErbB2 ubiquitylation and robust endocytic degradation.

438 439 440 441 442 443 444 445

0

3

6

9 hrs

40

er

at

in

ib

l tro on

IB: HSP90

ib

0

IB: ErbB2

tin

ErbB2 IP

N

IB: Tubulin

80

pa

55KD

IB: HSP90

120

La

IB: ErbB2

coIP quantitation Relative HSP90 (%)

C

on tr La ol pa ti N nib er at in ib

Input

150KD 116KD

150KD 116KD

ErbB2 transcription

40

C

Relative ErbB2 (%)

Neratinib GA GA+Neratinib 0 3 6 9 0 3 6 9 0 3 6 9 hrs IB:ErbB2 150KD IB:Tubulin 55KD

Fig. 7. Neratinib affects ErbB2 stability through unleashing HSP90 from ErbB2. (A, B) SKBR3 and AU565 cells were treated with neratinib (500 nM), the HSP90 inhibitor geldanamycin (1 μM), or both together for indicated times and lysed. Samples were analyzed by immunoblotting with ErbB2 antibody. Tubulin was used to confirm equal loading. The graphs below show quantification data of relative ErbB2 abundance during treatment time course. (C) SKBR3 cells were treated with DMSO (control), lapatinib (500 nM), or neratinib (500 nM) for 6 hours and lysed. ErbB2 was immunoprecipitated from cell lysates and samples were finally analyzed by immunoblotting with anti-ErbB2 and anti-HSP90 antibodies. Co-immunoprecipitated HSP90 band intensities were quantified and adjusted to the corresponding ErbB2 signal and shown in the bar chart.

429 430 431 432 433 434 435 436 437

ErbB2

Cell surface

Neratinib

B

418 419 420 421 422 423 424 425 426 427 428

9

neratinib treatment. In accordance with previous reports, proteasomes negatively regulate neratinib-induced ErbB2 endocytosis, likely through association with deubiquitylation activity [15]. Since HSP90 has been recognized as an important chaperone protein for many kinases including ErbB2 and RAF, we examined whether neratinib affected the interaction of ErbB2 with HSP90. Unlike lapatinib, neratinib triggered effective dissociation of HSP90 from ErbB2, as revealed by co-immunoprecipitation assays. This

observation is consistent with a previous study by Yarden and colleagues with the CI-1033 inhibitor (Canertinib) [18]. During the clinical management of HER2-positive breast cancers, the resistance to trastuzumab and lapatinib develops frequently, which can be caused by ErbB2 mutations. Given that neratinib can induce the endocytic degradation of ErbB2 and thus reduces overall ErbB2 expression, this inhibitor cannot only block signal transduction downstream of ErbB2 but also can reduce the possibility to acquire ErbB2 mutations. Although in our cell cycle and apoptosis assays, neratinib failed to display advantages over lapatinib with short term treatment, further investigations await to disclose the effects of neratinib with long term treatment. In the present study, we provide detailed descriptions of the effects of lapatinib and neratinib on ErbB2 levels. As described in our working model, lapatinib induces strong up regulation of ErbB2 mRNA but limited endocytosis, while neratinib moderately increases ErbB2 transcription while potently promoting endocytic degradation of ErbB2 (Fig. 8). Our findings reveal another aspect of the potential benefits of neratinib treatment in targeting ErbB2, which is also supported by recent investigations that demonstrate the efficacy of neratinib in overcoming trastuzumab resistance and its effectiveness in clinical evaluations [2,19,20].

446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468

Authors’ contributions ML and HL designed the project. YZ, JZ, CL, SD, LF, XL, YZ, YS, TW, and YW performed all experiments. YZ, JZ, WC, and SM analyzed the data. YZ, JZ, and HL wrote the manuscript. All authors reviewed the manuscript.

469 470 471 472 473 474 475

Acknowledgments

476 477 478 This work was supported by the National Natural Science Foundation of China [No. 81301901]; and the “Climbing Scholar” and Q6 479 480 “Excellent Talents” schemes of Liaoning Province, China.

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

ARTICLE IN PRESS 10

Y. Zhang et al. / Cancer Letters ■■ (2016) ■■–■■

481 Conflict of interest 482 483 The authors declare no conflict of interest. 484 485 Q7 References 486 [1] A. Citri, Y. Yarden, EGF-ERBB signalling: towards the systems level, Nat. Rev. 487 Mol. Cell Biol. 7 (7) (2006) 505–516. 488 [2] M. Segovia-Mendoza, et al., Efficacy and mechanism of action of the tyrosine 489 kinase inhibitors gefitinib, lapatinib and neratinib in the treatment of HER2490 positive breast cancer: preclinical and clinical evidence, Am. J. Cancer Res. 5 491 (9) (2015) 2531–2561. 492 [3] M.F. Rimawi, R. Schiff, C.K. Osborne, Targeting HER2 for the treatment of breast 493 cancer, Annu. Rev. Med. 66 (2015) 111–128. 494 [4] A. Wissner, et al., Synthesis and structure-activity relationships of 6,7495 disubstituted 4-anilinoquinoline-3-carbonitriles. The design of an orally active, 496 irreversible inhibitor of the tyrosine kinase activity of the epidermal growth 497 factor receptor (EGFR) and the human epidermal growth factor receptor-2 498 (HER-2), J. Med. Chem. 46 (1) (2003) 49–63. 499 [5] S.K. Rabindran, et al., Antitumor activity of HKI-272, an orally active, irreversible 500 inhibitor of the HER-2 tyrosine kinase, Cancer Res. 64 (11) (2004) 3958–3965. 501 [6] A. Wissner, T.S. Mansour, The development of HKI-272 and related compounds 502 for the treatment of cancer, Arch. Pharm. (Weinheim) 341 (8) (2008) 465–477. 503 [7] R. Avraham, Y. Yarden, Feedback regulation of EGFR signalling: decision making 504 by early and delayed loops, Nat. Rev. Mol. Cell Biol. 12 (2) (2011) 104–117. 505 [8] J. Baulida, et al., All ErbB receptors other than the epidermal growth factor 506 receptor are endocytosis impaired, J. Biol. Chem. 271 (9) (1996) 5251–5257. 507

[9] A. Sorkin, L.K. Goh, Endocytosis and intracellular trafficking of ErbBs, Exp. Cell Res. 315 (4) (2009) 683–696. [10] W. Xu, et al., Chaperone-dependent E3 ubiquitin ligase CHIP mediates a degradative pathway for c-ErbB2/Neu, Proc. Natl. Acad. Sci. U.S.A. 99 (20) (2002) 12847–12852. [11] N.M. Pedersen, et al., Geldanamycin-induced down-regulation of ErbB2 from the plasma membrane is clathrin dependent but proteasomal activity independent, Mol. Cancer Res. 6 (3) (2008) 491–500. [12] C. Marx, et al., ErbB2 trafficking and degradation associated with K48 and K63 polyubiquitination, Cancer Res. 70 (9) (2010) 3709–3717. [13] K. Tian, et al., Neuroleukin/autocrine motility factor receptor pathway promotes proliferation of articular chondrocytes through activation of AKT and Smad2/3, Sci. Rep. 5 (2015) 15101. [14] M.J. Clague, H. Liu, S. Urbe, Governance of endocytic trafficking and signaling by reversible ubiquitylation, Dev. Cell 23 (3) (2012) 457–467. [15] H. Liu, et al., Regulation of ErbB2 receptor status by the proteasomal DUB POH1, PLoS ONE 4 (5) (2009) e5544. [16] H. McDonough, C. Patterson, CHIP: a link between the chaperone and proteasome systems, Cell Stress Chaperones 8 (4) (2003) 303–308. [17] R.L. Siegel, K.D. Miller, A. Jemal, Cancer statistics, 2016, CA Cancer J. Clin. 66 (1) (2016) 7–30. [18] A. Citri, et al., Drug-induced ubiquitylation and degradation of ErbB receptor tyrosine kinases: implications for cancer therapy, EMBO J. 21 (10) (2002) 2407–2417. [19] A. Canonici, et al., Neratinib overcomes trastuzumab resistance in HER2 amplified breast cancer, Oncotarget 4 (10) (2013) 1592–1605. [20] C.L. Schwab, et al., Neratinib shows efficacy in the treatment of HER2 amplified carcinosarcoma in vitro and in vivo, Gynecol. Oncol. 139 (1) (2015) 112– 117.

Please cite this article in press as: Yingqiu Zhang, et al., Neratinib induces ErbB2 ubiquitylation and endocytic degradation via HSP90 dissociation in breast cancer cells, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.08.026

508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536