BBAMCR-17603; No. of pages: 11; 4C: 8 Biochimica et Biophysica Acta xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbamcr

3Q2

Yuliia Yuzefovych, Rainer Blasczyk, Trevor Huyton ⁎

4

Institute for Transfusion Medicine, Hannover Medical School, Carl-Neuberg-Str. 1, D-30625 Hannover, Germany

5

a r t i c l e

6 7 8 9 10

Article history: Received 2 April 2015 Received in revised form 17 June 2015 Accepted 19 June 2015 Available online xxxx

11 12 13 14 15

Keywords: ANP32C Hsp90 Cell cycle Cancer

a b s t r a c t

E

D

P

The acidic nuclear phosphoproteins (ANP32A-H) are an evolutionarily conserved family of proteins with diverse and sometimes opposing cellular functions. Here we show that the oncogenic family members ANP32C and ANP32D are associated in complexes containing the molecular chaperone Hsp90. The oncogenic ANP32C protein appears to be highly unstable with a rapid degradation (t1/2 N 30 min) occurring upon treatment of cells with cycloheximide. ANP32C was also found to be associated with oncogenic Hsp90 complexes by virtue of its ability to interact and be immunoprecipitated by the Hsp90 inhibitor PU-H71. Further studies treating cells with the Hsp90 inhibitors PU-H71 and 17-AAG showed atypical increased protein stability and prevention of ANP32C degradation compared to the Hsp90 client AKT. Cells overexpressing ANP32C or its mutant ANP32CY140H showed enhanced sensitivity to treatment with PU-H71 as demonstrated by CCK-8 and colony formation assays. Our results highlight that certain malignancies with ANP32C/D overexpression or mutation might be specifically targeted using Hsp90 inhibitors. © 2015 Published by Elsevier B.V.

31 29 28

C

30

1. Introduction

33 34

The ANP32 family of proteins [1,2] belongs to the superfamily of leucine-rich repeat (LRR) containing proteins [3]. The simple domain structure of ANP32 proteins is highly conserved and comprises a capped LRR motif at the N-terminus [4], a central domain (amino acids ~150– 174) containing a region mediating cellular transformation [5] and a highly acidic C terminus containing ~ 70% aspartic and glutamic acid residues. The ANP32A (pp32) protein is the most characterized family member and has been demonstrated to act as a tumor suppressor [6]. In contrast ANP32C (pp32r1) is oncogenic and found to be overexpressed in poorly differentiated tumors [5,7,8]. Expression of ANP32C also appears to be more predominant in common cancer cell lines than in normal fibroblasts [9]. Several functional mutations in ANP32C have been found to be associated with an increase in proliferation when transfected into cells [10, 11]. We have recently determined that the increase in proliferation upon overexpression of ANP32C or ANP32C_Y140H occurs through the dysregulation of Chromodomain Helicase DNA-binding protein 4 (CHD4) mediated cell cycle control [12,13] and appears to be confined to p53wt cells, since ANP32C and ANP32C_Y140H expressing cells with mutated p53 were not hyperproliferative [14]. The overexpression effects were also shown to increase resistance to the chemotherapeutic drug FTY720 although the specific mechanism of this resistance is still unclear [15]. Other recent studies have also suggested a possible role

43 44 45 46 47 48 49 50 51 52 53 54 55 Q3

R

R

N C O

41 42

U

39 40

E

32

37 38

16 17 18 19 20 21 22 23 24 25 26

T

27

35 36

O R O

i n f o

F

2

Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90

1Q1

⁎ Corresponding author. E-mail address: [email protected] (T. Huyton).

for ANP32C in the cytoplasmic degradation of Human antigen R (HuR) in oral carcinoma cell lines [9]. In contrast however we have not observed any interaction of overexpressed ANP32C with HuR in kidney cancer (ACHN) or cervical carcinoma (HeLa) cell lines suggesting that some ANP32C functional differences may occur between different cancer cell types [15]. In this work we sought to further explore the interactome of ANP32C and identified Hsp90 as an ANP32C interacting protein. Hsp90 is an evolutionary conserved protein chaperone and represents one of the most highly expressed cellular proteins. The function of Hsp90 is essential for maintaining proteome homeostasis through the control of correct folding, stability and functional activity, intracellular allocation and protein turnover under normal and stress-inducing conditions. Hsp90 is an ATPase with working cycle that includes multiple conformational changes triggered by ATP binding and hydrolysis as well as successive interactions with different partners, so-called “co-chaperones” (e.g. Hsp70, HOP, Cdc37, p23) [16]. A subset of cellular proteins depending on HSP90's chaperoning activity and usually referred as ‘clients’ accounts for 10–30% of the proteome [17,18]. Kinases, E3 ligases and transcription factors represent major classes among these Hsp90 client proteins [19]. Many of these client proteins play key roles as effectors in signal transduction pathways regulating cellular proliferation, cell cycle control, division, DNA-damage response and apoptosis. They are also when mutated or overexpressed often found to be involved in malignant transformation (i.e., p53, Erb2, EGFR, v/c-Src, AKT, Raf-1) [20]. The ability to protect mutated, destabilized or alternately expressed proteins confers a balancing role on Hsp90 that can promote the activity of certain oncoproteins facilitating cancer cell survival and tumor

http://dx.doi.org/10.1016/j.bbamcr.2015.06.007 0167-4889/© 2015 Published by Elsevier B.V.

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110

2. Materials and methods

112

2.1. Lentiviral constructs for ANP32 family proteins

113 114

117

The ANP32A, B, D and E full-length genes were amplified by PCR from full-length cDNA clones (Source Biosciences lifescience). The fragments were then first cloned as previously described for the ANP32C gene into the pCDNA-V5-His-TOPO vector (Invitrogen) before subcloning into the pRRL.PPT.SFFV.mcs.pre vector [14].

118

2.2. Cell culture and lentiviral transduction

119 120

ACHN cells (DSMZ) were grown at 37 °C in a 5% CO2-humidified atmosphere in Dulbecco's modified Eagle's medium (DMEM) supplemented with 50 U/ml penicillin, 50 μg/ml streptomycin, 2 mM L-glutamine, 1% NEAA, 1 mM Sodium Pyruvate and 10% fetal bovine serum (FBS). HeLa and HEK293T cells were grown in DMEM supplemented with; 50 U/ml penicillin, 50 μg/ml streptomycin, 2 mM L-glutamine. Lentivirus was produced in HEK293T cells using standard protocols and ACHN or HeLa cells were transduced as previously described [14].

126 127

128 129 130 131 132 133 134 135 136 137 138 139 140

2.5. Peptide mass fingerprinting

149

The gel bands were cut into small pieces, de-stained with acetonitrile and washed repeatedly (8 M urea/ 0.4 M NH4HCO3). Samples were then reduced with dithiothreitol for 30 min at 55 °C and alkylated with iodoacetamide for 15 min at room temperature in the dark. Subsequently the gel pieces were washed twice and shrunken with acetonitrile. The gel pieces were then covered with mass-spec grade trypsin (Serva, porcine) in cleavage buffer (2 M urea/0.1 M NH4HCO3). The digest was performed at 37 °C overnight. From each sample 1 μl of the supernatant after digest was used for the MALDI-MS measurement. The MALDI-MS data were used for a database search with the software Mascot (Matrix Science) using the SwissProt database, species: human. Peptide mass tolerance was set to 100 ppm.

150

2.6. Co-immunoprecipitation

162

2.3. Fractionation of ANP32C from overexpressing cell lines by size exclusion chromatography (SEC) For “Native” fractionation of lysates HeLa cells overexpressing ANP32C or ACP32CY140H were grown in T150 flasks until ~ 90% confluency. Cells were then washed 3 times with PBS before being scraped from the flask surface in PBS containing 1000 U benzonase. Cells were then mechanically lysed by repeatedly passing the cell suspension through a fine 27 gauge needle. The cell lysate was then centrifuged at 13,000 rpm for 10 min to remove cell debris and the supernatant loaded onto a superdex 200 16/60 gel filtration column equilibrated in PBS, collecting 0.5 ml fractions. The elution profile was compared to standards of Blue Dextran (Mw ~ 2,000,000 Da), Rabbit IgG (Mw ~150,000 Da) and Bovine serum Albumin (Mw ~66,000 Da).

144 145 146 147 148

151 152 153 154 155 156 157 158 159 160 161

For co-immunoprecipitation (Co-IP) experiments cells were lysed on ice by scraping and 30 min incubation in lysis buffer containing 1% NP40, protease inhibitors cocktail (Roche) and Benzonase (Merck). In some experiments 20 mM sodium molybdate was added to the lysis buffer. Protein quantitation was performed using a BCA assay (Uptima). Lysates were precleared with protein A sepharose (GE Healthcare) before a total of 1,5-3,5 mg protein was immunoprecipitated by antiV5 tag (Serotech, MCA1360) antibody pre-immobilized on protein A beads. Beads were then washed with lysis buffer and PBS or phosphate buffer (pH 7,4) with high salt (0,5 M NaCl) and/or 20 mM sodium molybdate where necessary. Proteins were eluted in loading dye and resolved on 4–20% SDS-PAGE gels (Invitrogen) and analyzed by western blotting with anti-Hsp90 (Abcam, ab178854) or anti-Hsp70 (Thermo Scientific, MA1-90504) antibodies. Co-IP for detection of the interaction between ANP32CY140H and Hsp70 was preceded by crosslinking of anti-V5 tag antibody to the beads utilizing DMP (dimethyl-pimelimidate, Thermo Fisher) crosslinker. For serial IPs a portion of HeLa C lysate with 250 μg of total protein (prepared as for a standard IP) was incubated with anti-V5 or antiHsp90 antibody-coupled beads in 4 successive steps. Equal aliquots of the remaining supernatant were collected after each step and analyzed by western blotting with anti-V5 tag or anti-Hsp90 antibody.

163

2.7. Cycloheximide chase assay

186

HeLa C/YH cells were plated at a density of 3 × 105 (or 2 × 105) cells per well of 6-well plate 16 h before treatment with 50 μg/ml CHX (0.5 h, 1 h, 2 h, 3 h) or 0.05% DMSO (3 h) as a vehicle control. ACHN WT-D cells were treated with 50 μg/ml CHX or 0.05% DMSO as a vehicle control for 3 h. Cells were collected by scraping in RIPA buffer and lysed for 30 min on ice. Samples were resolved on 4–20% SDS-PAGE gels and analyzed by western blotting with anti-V5 tag antibody. For combined CHX and PU-H71 (Cayman Chemicals) treatment HeLa C/YH cells were plated as described above and treated first for 8 h (or 24 h) with 1 μM PU-H71 and then with both 1 μM PU-H71 and 5 μg/ml CHX for 0.5 and 1 h. Cells were lysed as described before. For the Co-IP with newly translated ANP32CY140H HeLa C/YH cells at 80% confluence were treated with 50 μg/ml CHX for 3 h. After the

187 188

T

C

E

R R

O

C

125

N

123 124

U

121 122

143

E

111

115 116

The cell permeable and thiol-cleavable crosslinker Dithiobissuccinimidyl propionate (DSP) (Thermo Fisher) was used according to the protocol of Smith et al. [30] with minor modifications. Cells were crosslinked in 2 mM DSP for 30 min at room temperature and lysed in NP-40 buffer containing protease inhibitors cocktail (Roche) and Benzonase (Merck).

F

90

141 142

O

88 89

2.4. Cross-linking and immunoprecipitation of ANP32C from overexpressing cell lines

R O

86 87

progression [21]. This is consistent with the observation of Hsp90 upregulation in tumors in comparison to normal tissues [22,23], emphasizes the importance of Hsp90 for tumor cell growth and highlights Hsp90 as an attractive target for cancer treatment. The inhibition of Hsp90 makes it possible to affect a number of oncogenic pathways that are known to be under control of the chaperone. Hsp90 found within tumor cells has been suggested to be composed of two distinct pools; Hsp90 that participates in formation of large heterogenous protein complexes that are especially abundant in oncogenic conditions and uncomplexed Hsp90 that can be found in all cells, regardless of their oncogenicity [24]. This original work was however refuted in subsequent work by another group [25], however certain specific and potentially “oncogenic” Hsp90 complexes do exhibit higher affinity to certain Hsp90 inhibitors, in particular PU-H71 [26,27]. Many small molecules have now been discovered and chemically synthesized to target Hsp90 in cancer. Almost all Hsp90 inhibitors currently undergoing clinical investigation block the chaperone ATPase function, leading to degradation of the oncogenic client and antitumor activity [28,29]. Of these with the geldanamycin derivative 17-AAG is among those that have been tested in clinical studies. The promising effectiveness of Hsp90 inhibitors in different malignancies together with growing body of evidence on Hsp90's biology keeps Hsp90 in the focus of researchers' attention and makes Hsp90 inhibition a highly attractive therapeutic strategy for cancer treatment. In this work we describe for the first time the association between Hsp90 and the oncogenic ANP32 family members ANP32C and ANP32D and the effect that this has on the efficacy of Hsp90 inhibitors.

D

84 85

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

P

2

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185

189 190 191 192 193 194 195 196 197 198 199

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

210

2.9. Western blotting

211 212

221

The transfer was performed using iBlot Dry Blotting System (Invitrogen). After transfer a polyvinylidene difluoride (PVDF) membrane was blocked for 1 h in 5% low fat milk powder in PBST. The membrane was then incubated overnight at 4 °C with primary antibodies. The blot was then washed 3 times with PBST followed by incubation with 1:2500 dilution of rabbit-anti-mouse-HRP (DAKO) or goat-antirabbit-HRP (DAKO) conjugated secondary antibody for 1 h at room temperature. Following 3 washes with PBST detection was performed by incubating the blot directly with Clarity western ECL substrate (Bio-Rad) and measured on a MultiImage II chemiluminescence imaging system (Alpha Innotech).

222

2.10. CCK-8 assay

223 224

234 235

CCK-8 assay is a colorimetric method for the determination of the number of viable cells in cell proliferation and cytotoxicity studies. It utilizes WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4disulfophenyl)-2H-tetrazolium, monosodium salt] that gets reduced by dehydrogenases in cells to give an orange colored soluble product WST-8-formazan. The amount of the formazan dye generated by dehydrogenases in cells is directly proportional to the number of living cells and can be measured spectrophotometrically at 450 nm. For the CCK-8 assay cells were plated in 96-well plates at 3000 per well ~36 h before treatment. Treatment with different concentrations (50, 100, 125, 250, 500, 750, 1000 nM) of PU-H71 lasted for 30 h before CCK-8 reagent was added for 1 h and colorimetric reaction was measured on Synergy 2 Microplate Reader (BioTek).

236

2.11. Colony growth assay

237

242

Cells were plated in 6-well plates at a density of 3000 cells per well. After 24 h cells were treated with 50 nM or 1 μM PU-H71 for 30 h. 0.1% DMSO was used as a vehicle control. After the incubation with drug cells were washed with PBS and then grown in fresh media with regular media exchange until colonies were visible. The plates were fixed and stained with methylene blue solution in 50% methanol.

243 244

2.12. Cell treatment with PU-H71 and 17-AAG for WB, time-course and Co-IP

245

Cells were plated out one day before treatment at a concentration to reach 70–80% confluence at the moment of treatment. Treatment was performed with 1 μM PU-H71 and 1 μM 17-AAG for 24 h. Cells were collected by scraping in RIPA buffer and lysed for 30 min on ice. Samples were resolved on 4–20% SDS-PAGE gels and analyzed by western blotting with anti-V5, Hsp90, Hsp70, GAPDH (Santa Cruz, 0411 sc-47724) and pan-AKT (Cell Signaling, 40D4 #2920) antibodies. PU-H71 time-course was performed on HeLa C/YH cells plated at a density of 3 × 105 cells per well of 6-well plate 16 h before treatment. 1 μM PU-H71 was added for 0.5 h, 1 h, 1.5 h, 3 h, 5 h, 8 h or 24 h. 0.1%

229 230 231 232 233

238 239 240 241

246 247 248 249 250 251 252 253 254

T

227 228

C

225 226

E

219 220

R

217 218

R

215 216

N C O

213 214

U

207 208

3. Results

260

HeLa cells ectopically expressing the ANP32C protein or its mutant ANP32C_Y140H were generated by lentiviral transduction as described in the Materials and methods and are subsequently referred to as HeLa C and HeLa YH in the text.

261

3.1. ANP32C is found in high molecular weight cytoplasmic complexes

265

Lysates of native ANP32C protein complexes were prepared by resuspending cells in PBS containing protease inhibitors and subjecting samples to mechanical lysis by repeatedly passing the lysate through a small gauge needle. The resulting native protein complexes were then fractionated by size/shape using size exclusion chromatography (SEC) with collection of sequential fractions. Protein standards were analyzed using the same SEC method to identify the approximate molecular weight range across each fraction. ANP32C containing fractions were identified via western blotting using an anti-V5 antibody and illustrated that ectopically expressed ANP32C exists in a large range of high molecular weight protein complexes (Fig. 1a). To stabilize these ANP32C containing complexes we utilized the Reversible Cross-Link ImmunoPrecipitation (ReCLIP) approach using the cell permeable and thiol-cleavable crosslinker Dithiobis-succinimidyl propionate (DSP) [30]. Western blotting of cell lysates from HeLa C or HeLa YH cells treated with DSP was compared under reducing and non-reducing conditions. In non-reducing conditions the stabilization of several high molecular weight complexes ~90 and ~160 kDa containing ANP32C could be achieved (Fig. 1b).

266 267

F

209

Chemical precipitation experiments with PU-H71-conjugated beads on lysates of HeLa WT, HeLa C and HeLa YH cells were performed as described by Taldone et al. [31]. PU-H71 coupled beads were generously donated by Prof. Gabriella Chiosis (Memorial Sloan Kettering Cancer centre, USA).

O

205 206

R O

2.8. Chemical precipitation with PU-H71-conjugated beads

255 256

P

204

DMSO served as a vehicle control. Cells were lysed as described previously. For the Co-IP HeLa C/YH cells at 80% confluence were treated with 1 μM PU-H71 or DMSO (0.1%) as a vehicle control for 24 h. Cell lysis was performed in a standard for IP way.

D

202 203

incubation the CHX-containing media was withdrawn, cells were washed with PBS and fresh media was used. Cells were incubated in the fresh media for 30 min. Lysis was performed in a standard for IP way.

E

200 201

3

257 258 259

262 263 264

268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284

3.2. Mass spectrometry identification of proteins in complexes containing 285 ANP32C 286 To isolate intact ANP32C containing complexes cells were again treated with DSP and subjected to immunoprecipitation using protein-A beads coupled with anti-V5 antibody. Comparison of control lysates and ANP32C expressing lysates, by SDS PAGE under reducing conditions, illustrated several specific protein bands which were excised from the gel and subjected to peptide mass fingerprinting (Fig. 1c). A number of proteins were identified from these bands including Heat shock protein 90A/B, Tubulin, Heat shock cognate 71 kDa protein and Heat shock 70 kDa protein 1A/1B, full details are provided in Table 1 and Supplementary Fig. 1.

287 288

3.3. Overexpression of ANP32C does not modulate Hsp90 or Hsp70 levels

297

We first sought to determine if our cell lines expressing ANP32C/ Y140H showed increased levels of Heat shock proteins. Western blotting demonstrated consistent levels of both Hsp90 and Hsp70 in wild type and expressing cell lines (Fig. 2a).

298 299

3.4. Confirmation of ANP32C in complexes containing Hsp90 and Hsp70

302

To confirm the presence of ANP32C in Hsp90 or Hsp70 containing complexes we performed co-immunoprecipitation experiments on untreated cell lysates. The ANP32C protein was pulled down with proteinA beads coupled with anti-V5 antibody and washed extensively before blotting with anti-Hsp90 or anti-Hsp70. As can be seen in Fig. 2b and Supplementary Fig. 2, ANP32C and ANP32C_Y140H co-precipitated with large amounts of Hsp90. Likewise co-immunoprecipitation of Hsp70 with ANP32C/Y140H could also be observed (Fig. 2c).

303

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

289 290 291 292 293 294 295 296

300 301

304 305 306 307 308 309 310

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

D

P

R O

O

F

4

318 319 320 321

t1:1 t1:2

E

R

R

316 317

O

315

C

313 314

Native protein fractionation using size exclusion chromatography (SEC) with collection of sequential fractions was also repeated and fractions evaluated by western blotting for ANP32C/Y140H and Hsp90. As can be seen in Supplementary Fig. 3 significant co-elution of ANP32C/ Y140H and Hsp90 could be observed in SEC fractions, however we could also demonstrate that not all ANP32C protein is complexed with Hsp90 through sequential immunoprecipitation of V5 tagged ANP32C and blotting Hsp90 in the remaining supernatant (Supplementary Fig. 4). The reverse experiment also showed that immunoprecipitation of all Hsp90 was not sufficient to deplete all V5 tagged ANP32C from the remaining supernatant (Supplementary Fig. 4).

Table 1 Peptide mass fingerprinting.

N

311 312

C

T

E

Fig. 1. ANP32C is found in high molecular weight cytoplasmic complexes: Size exclusion chromatography of HeLa YH expressing lysates and analysis of fractions by western blot with antiV5 antibody demonstrates a broad elution profile from 2000 to 60 kDa and is eluted as two major peaks. b) Crosslinking of ANP32C/Y140H proteins using the reversible crosslinker DSP: Lysates from cells treated with DSP were compared under reducing (R) and non-reducing (NR) conditions before western blotting with anti-V5 antibody. Cross-linked complexes are observed which resolve at ~90 and 160 kDa. c) Mass spectrometry identification of proteins in complexes containing ANP32C: Immunoprecipitated cross-linked complexes were isolated and resolved by standard SDS PAGE. Compared to HeLa cell lysates cross-linked HeLa YH lysates demonstrated additionally bound proteins. Regions 1–4 were excised and identified by mass spectrometry (see Table 1).

Protein band

Identified proteins

Mascot (Protein score)

Sequence coverage

t1:4

Band-1

t1:5

Band-2

t1:6

Band-3

HS90A_HUMAN HS90B_HUMAN HSP7C_HUMAN HSP71_HUMAN TBB5_HUMAN TBA1B_HUMAN TBB4B_HUMAN TBB2A_HUMAN TBB4A_HUMAN TBB3_HUMAN TBA1A_HUMAN TBA1C_HUMAN TBA3C_HUMAN ND

149 272 175 62 153 94 141 131 108 73 67 67 51 ND

38% 46% 39% 20% 56% 50% 54% 52% 48% 28% 43% 43% 38% ND

t1:7 t1:8

Band-4

ND (not determined).

U

t1:3

3.5. ANP32C behaves as a client of Hsp90

322

As discussed in the Introduction the Hsp90 chaperone forms interactions with both co-chaperone molecules and client proteins many of which are listed in the Hsp90 database (http://www.picard.ch/ downloads/downloads.htm). These interactions can be distinguished by co-immunoprecipitation under specific chemical conditions namely the presence of molybdate ions to stabilize the Hsp90:client/ co-chaperone complexes and high salt washes to remove cochaperone molecules since molybdate characteristically enforces salt resistance of Hsp90:client complexes [32–34]. We therefore performed immunoprecipitation experiments to determine if ANP32C behaves as a co-chaperone or client protein. We observed that high salt washes had little effect, however the presence of molybdate ions significantly enhanced the stability of ANP32C/Y140H:Hsp90 containing complexes suggesting that ANP32C is a newly identified client protein of Hsp90 (Fig. 2d).

323

3.6. ANP32C is rapidly degraded by the proteasome

338

One of the important functions of Hsp90 chaperoning is to mediate the degradation of its various client substrates through the proteasomal degradation machinery. We therefore performed experiments using the protein translation inhibitor cycloheximide and proteasome inhibitor MG132 to determine the stability of ANP32C proteins. As depicted in Fig. 3a–b, ANP32C and ANP32C/Y140H proteins were shown to be extremely unstable and N50% of the protein was degraded within 30 min. In contrast treatment with the proteasomal inhibitor MG132 clearly enhanced the stability of ANP32C and its mutant proteins

339 340

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

324 325 326 327 328 329 330 331 332 333 334 335 336 337

341 342 343 344 345 346 347

5

P

R O

O

F

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

348

C

T

E

D

Fig. 2. a) Overexpression of ANP32C and ANP32C_Y140H does not change levels of Hsp90 and Hsp70: Western blot analysis to detect expression of Hsp90 and Hsp70 was performed on lysates of HeLa cells expressing V5-tagged ANP32C and ANP32C_Y140H cells (HeLa C and HeLa YH cells respectively) in comparison to wild type HeLa cells. V5 blot shows expression of ANP32C/Y140H. GAPDH served as loading control. b) Interaction of ANP32C/Y140H with Hsp90: Co-immunoprecipitation (Co-IP) was performed with anti-V5 antibody to capture V5tagged ANP32C or ANP32C_Y140H proteins expressed in HeLa C and HeLa YH cells and blotted with antibody against Hsp90. Wild type HeLa cells served as a control. Representative results are shown, including V5 blot of Co-IP samples and Hsp90 input. c) Interaction of ANP32C/Y140H with Hsp70: Co-IP was performed with anti-V5 antibody to capture V5-tagged ANP32C/ Y140H proteins from HeLa C and HeLa YH cells and blotted with antibody against Hsp70. Wild type HeLa cells served as a control. Representative results are shown, including V5 blot of CoIP samples and Hsp70 input. d) Sodium molybdate stabilizes interaction of ANP32C/Y140H with Hsp90 under high salt conditions: Co-IP was performed with anti-V5 antibody to capture V5-tagged ANP32C/Y140H proteins from HeLa C and HeLa YH cells and blotted with antibody against Hsp90. Lysates for each cell type were prepared either with 20 mM sodium molybdate (3, 4, 7, 8) or without this agent (1, 2, 5, 6). Co-IP samples of regular or sodium molybdate containing lysates were washed either with low salt (137 mM NaCl) buffer (1, 3, 5, 7) or high salt (500 mM NaCl) buffer (2, 4, 6, 8). Representative results are shown, including V5 blot of Co-IP samples and Hsp90 input (M corresponds to molybdate).

and led to the accumulation of ubiquitinated ANP32C proteins (Supplementary Fig. 5a,b).

350

3.7. Newly translated ANP32C protein is rapidly associated with Hsp90

351 352

358

To demonstrate that reduced levels of ANP32C protein are also readily bound by Hsp90 complexes we performed a pulse-chase immunoprecipitation experiment. First translation of ANP32C mRNA was inhibited by cycloheximide for 3 h, as shown previously in Fig. 3b–c, to be sufficient to allow complete degradation of ANP32C/Y140H proteins. Cells were then left to recover for 30 min before immunoprecipitation. These newly translated proteins were again shown to be readily bound by Hsp90 (Fig. 3c).

359

3.8. Only ANP32C and ANP32D are novel clients of Hsp90

360 361

ANP32C belongs to a family of proteins all sharing high sequence identity and in some cases overlapping functional roles in cellular processes. The observation that ANP32C formed complexes containing Hsp90 and behaves as a client protein therefore prompted us to determine if other ANP32 family members were also clients of Hsp90. Coimmunoprecipitation experiments on cell lysates from ACHN cells expressing ANP32A, B, C, C-YH, D and E were performed in parallel. As can be seen from Fig. 3d, only ANP32C and ANP32D proteins were found to co-precipitate in complexes containing Hsp90. Similar to ANP32C the presence of molybdate ions significantly enhanced the stability of ANP32D:Hsp90 containing complexes that also showed high-salt resistance suggesting that ANP32D is also a client protein of Hsp90 (Supplementary Fig. 6).

356 357

362 363 364 365 366 367 368 369 370 371 372

R

R

N C O

354 355

U

353

E

349

We then repeated cycloheximide inhibition studies on cell lines expressing all the ANP32 family members. In these experiments we observed that only the oncogenic family members ANP32C and ANP32D were intrinsically unstable and rapidly degraded, while the other family members showed no sign of degradation (Fig. 3e).

373

3.9. ANP32C is co-precipitated with “oncogenic” Hsp90 complexes

378

Proteomics data has shown that multiple oncoproteins involved in tumor proliferation, survival and invasive potential are found in complex with PU-H71-bound Hsp90 [26]. We performed chemical precipitation using PU-H71 coupled beads (generously donated by Prof. Gabriella Chiosis, Memorial Sloan Kettering Cancer centre). In control HeLa cell lysates Hsp90 could be efficiently pulled-down, while in ANP32C/Y140H expressing lysates complexes containing ANP32C/ Y140H:Hsp90 were pulled down (Fig. 4a).

379

3.10. Atypical effects of PU-H71 inhibition on ANP32C degradation

387

It has been demonstrated that many clients of Hsp90 are degraded upon Hsp90 inhibition. To examine if this holds true for ANP32C/ Y140H we performed a time course experiment treating HeLa C/YH with PU-H71 for different time periods (Fig. 4b). This experiment showed a complex time-dependent response of ANP32C/Y140H protein level to treatment with PU-H71 with degradation within first 1.5 h of treatment followed by a gradual increase of protein levels. The observed degradation of ANP32C/Y140H supports the claim that ANP32C and its mutant are novel clients of Hsp90 folding machinery. The ANP32C/ Y140H degradation is characterized by fast kinetics with substantial drop in protein level already after 1.5 h of PU-H71 treatment and

388 389

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

374 375 376 377

380 381 382 383 384 385 386

390 391 392 393 Q4 394 395 396 397 398

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

E

C

T

E

D

P

R O

O

F

6

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

being only a first “early-response” phase in the response to Hsp90 inhibition. Starting from 3 h of treatment with PU-H71 we observed a “late-response” with gradually increased ANP32C/Y140H levels that is a nontrivial effect of Hsp90 inhibition on its clients. To further study this phenomenon of ANP32C/Y140H stabilization under conditions of prolonged presence of Hsp90 inhibitor PU-H71 we utilized another well-studied Hsp90 inhibitor 17-AAG. We treated cells with PU-H71 or 17-AAG for 24 h to monitor the levels of Hsp90, Hsp70, AKT and ANP32C and its mutant proteins compared to untreated controls (Fig. 4c). While the level of Hsp90 remained constant in treated or untreated samples the levels of Hsp70 are significantly upregulated as has been previously described [35]. As a positive control we performed western blotting of the well characterized Hsp90 client AKT and demonstrated significant degradation of AKT protein upon 24 h treatment with either 17-AAG or PU-H71. In contrast moderate increases in the levels of ANP32C/Y140H could be observed in treated samples (Fig. 4c), these quantified over multiple experiments (n = 5) were shown to be significant (p b 0.05) for HeLa C treated cells and HeLa YH cells treated with 17-AAG, but not HeLa YH treated with PU-H71 (Supplementary

N

400 401

U

399

C

O

R

R

Fig. 3. a) Cycloheximide (CHX) chase assay to determine stability of ANP32 family members: ACHN C cells were treated with CHX (50 μg/ml) for indicated periods of time and western blotting was performed. Treatment with DMSO as a vehicle control (0.05%) lasted 3 h. b) Cycloheximide chase assay to determine stability of ANP32 family members: ACHN YH cells were treated with CHX (50 μg/ml) for indicated periods of time and western blotting was performed. Treatment with DMSO as a vehicle control (0.05%) lasted 3 h. c) Newly synthesized ANP32C/Y140H interact with Hsp90: Co-IP was performed with anti-V5 antibody to capture V5-tagged ANP32C/Y140H proteins produced by HeLa C or HeLa YH cells within 0.5 h after 3 h-treatment with 50 μg/ml CHX and blotted with antibody against Hsp90. Wild type HeLa cells served as a control. Representative results are shown, including V5 blot of Co-IP samples and inputs for Hsp90 and V5. d) Studying of the interaction of ANP32 family members with Hsp90: Co-IP was performed with anti-V5 antibody to capture V5-tagged proteins of ANP32 family (ANP32A, ANP32B, ANP32C, ANP32C140YH, ANP32D, ANP32E) ectopically expressed in ACHN cells (ACHN A, ACHN B, ACHN C, ACHN YH, ACHN D and ACHN E respectively) and blotted with antibody against Hsp90. Wild type ACHN cells served as a control. Representative results are shown, including V5 blot of Co-IP samples and Hsp90 input. e) Cycloheximide chase assay to determine stability of ANP32 family members: ACHN A, B, C, YH, D and E cells were treated with CHX (50 μg/ml) or DMSO as a vehicle control (0.05%) for 3 h and western blotting was performed.

Fig. 7). These relatively small increases could however be further enhanced by combining a PU-H71 or 17-AAG pretreatment with a short-term cycloheximide incubation (Fig. 4d and Supplementary Fig. 8). Using equal amounts of PU-H71 treated and untreated samples we also compared the amount of Hsp90 that could be coimmunoprecipitated with ANP32C/Y140H proteins and observed that after prolonged incubation of cells with PU-H71 cells could coprecipitate with significantly more Hsp90 (Fig. 4e). This further highlights that long-term Hsp90 inhibition also inhibits degradation of ANP32C. To better understand the role of Hsp90 and its inhibition in degradation of ANP32C/Y140H we extended our studies with the proteasome inhibitor MG132. We found that combined treatment with MG132 and PU-H71 did not additionally affect the stabilization of ANP32C/ Y140H. This implies that the stabilizing effect of PU-H71 on ANP32C/ Y140H is exerted via the inhibition of proteasomal degradation and therefore supports the idea that apart from assistance in folding, Hsp90 can mediate proteasomal degradation of ANP32C and its mutant (Supplementary Fig. 5c).

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436

7

N C O

R

R

E

C

T

E

D

P

R O

O

F

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

U

Fig. 4. a) PU-H71 precipitation of Hsp90 complexes with ANP32C or ANP32C_Y140H: Chemical precipitation was performed with PU-H71-conjugated or control beads on lysates from HeLa C and HeLa YH cells and blotted with antibody against V5-tag. Wild type HeLa cells served as a control. Representative results are shown, including Hsp90 blot of chemical precipitation samples as well as Hsp90 and V5 input. b) PU-H71 time course: HeLa C/YH cells were treated with 1 μM PU-H71 for indicated periods of time or DMSO (0.1%) as a vehicle control for 24 h. ANP32C/Y140H protein levels were analyzed by western blotting with anti-V5 tag antibody. β-Actin served as a loading control. c) Hsp90 inhibitors PU-H71 and 17-AAG cause an increase in ANP32C/Y140H level: Western blot analysis to detect expression of Hsp90, Hsp70, AKT and ANP32C/Y140H was performed on lysates of wild type HeLa (1,4,7,10), HeLa C (2,5,8,11) and HeLa YH (3,6,9,12) cells treated with 1 μM PU-H71/17-AAG or DMSO (0.1%) as a vehicle control for 24 h. V5 blot shows expression of ANP32C/Y140H. GAPDH served as a loading control. d) Combined treatment with PU-H71 and CHX: HeLa YH cells were pretreated with 1 μM PU-H71 or DMSO (0.1%) as a control for 8 or 24 h before CHX at 5 μg/ml was added. Treatment with CHX lasted 0.5 or 1 h. Levels of ANP32C_Y140H and pan-AKT were analyzed by western blotting with anti-V5 and anti-pan-AKT antibodies respectively. GAPDH served as a loading control. e) PU-H71 stabilizes complexes of ANP32C/Y140H:Hsp90: HeLa C and HeLa YH cells were treated with 1 μM PU-H71 or DMSO (0.01%) as a vehicle control for 24 h. Co-IP was performed with anti-V5 antibody to capture V5-tagged ANP32C/Y140H proteins from PU-H71 or DMSO-treated HeLa C or HeLa YH cells and blotted with antibody against Hsp90. Representative results are shown, including V5 blot of Co-IP samples and Hsp90 input.

437

3.11. ANP32C expression confers sensitivity to PU-H71

438

To determine the dose dependent effect of PU-H71 treatment (50–1000 nM) on cell viability we carried out CCK-8 proliferation and colony formation assays. Cells overexpressing ANP32C or ANP32C_Y140H were visibly more sensitive than wild-type HeLa cells and showed more signs of apoptosis; detachment, shrinkage and rounding. These effects were clearly evident in CCK-8 measurements of cell proliferation (Fig. 5a). Additional colony formation assays were carried out at two concentrations (50 nM and 1 μM PU-H71) for

439 440 441 442 443 444 445

the three cell lines as shown in Fig. 5b. At these concentrations no colonies could be observed in ANP32C/Y140H expressing cells treated with 1 μM PU-H71, while wild-type cells retained a strong colony forming capacity.

446

4. Discussion

450

447 448 449

Our aims for this work were to explore in more detail the func- 451 tion of the ANP32C protein by identifying the proteins that interact 452 with it and help mediate the proliferative cellular phenotype that 453

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

8

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500

E

D

P

amount within 0.5 h while full degradation of the proteins could be seen after 1–2 h (Fig. 3b–d). Experiments using MG132 to block the proteasome, illustrated that the level of ANP32C/Y140H was dramatically increased, similar to the p53 positive control (Supplementary Fig. 5a). The ubiquitination of ANP32C/Y140H was also detected upon treatment with MG132 via co-immunoprecipitation experiments using anti-V5 and anti-ubiquitin antibodies (Supplementary Fig. 5b). Understanding the important role played by Hsp90 in both cellular pathologies and in normal cells is critical in developing new drugs that might be used to treat cancer and neurodegenerative diseases. There appears to be no known sequence consensus or structural elements within Hsp90 client proteins that enable us to distinguish a client from the rest of proteome [19]. Therefore to identify which proteins constitute the Hsp90 interactome and which specific cellular pathways are subsequently affected many groups have undertaken large scale proteomics based studies [19,26]. From these and other individual data an extensive list of Hsp90's client proteins and co-chaperones has been collated by Dr. Didier Picard (http://www.picard.ch/downloads/ downloads.htm) and presumably due to their low endogenous expression ANP32C or ANP32D are not found on this list. Hsp90 inhibitor based proteomic studies or inhibitor pull downs might also be unlikely to identify such “rapidly degraded” proteins like ANP32C/D. Similarly despite extensive screening of the Hsp90 interactome in yeast [19] certain proteins like the ANP32 family members have no attributed yeast homologue, highlighting that inter-species comparisons are not always applicable. It has been suggested that Hsp90 along with Hsp70 can participate in both folding and degradation for the same client even under physiological conditions [35] and we believe that this is likely to be the case for ANP32C and ANP32D. We therefore continued to explore the ANP32C:Hsp90 interaction by investigating the effects of Hsp90 ATP binding site specific drugs including 17-AAG, a synthetic variant of geldanamycin, and the purine scaffold inhibitor PU-H71. Using a bead conjugated Hsp90 inhibitor PU-H71 we were able to readily precipitate Hsp90 complexes containing ANP32C and its mutant (Fig. 4a). The majority of well-studied Hsp90 clients such as AKT and EGFR [26] or FANCA [39] show disruption of their interaction with Hsp90 and are rapidly degraded by the proteasome pathway upon cellular treatment with 17-AAG or PU-H71. Similarly we observed quick degradation of ANP32C/Y140H within 1.5 h after addition of PU-H71 during time course experiments. These effects of Hsp90 inhibition, although not sustained, are typical for classic Hsp90 clients and is an additional evidence for ANP32C/Y140H being an Hsp90 client (Fig. 4b). Prolonged treatment with PU-H71 leads to the gradual accumulation of ANP32C/ Y140H due to increased resistance to proteasomal degradation (Fig. 4c and Supplementary Fig. 5c). Treatment for 24 h with PU-H71 or 17-AAG showed moderately increased protein levels in comparison

T

C

467 468

E

465 466

R

463 464

R

461 462

O

459 460

C

457 458

we and others have previously described [10,14,15]. The only described ANP32C interacting protein so far is the Human antigen R (HuR) in oral carcinoma cell lines [9] and although we have failed to reproduce this interaction in either cervical (HeLa) or kidney (ACHN) cancer cell lines [15] however it is possible that other cell specific factors such as specific mRNAs might be required to mediate this interaction. Here we show for the first time that two members of the ANP32 family (ANP32C and ANP32D) are predominantly associated in high molecular weight complexes containing the cellular chaperones Hsp90 and Hsp70. The clients of Hsp90 are functionally and structurally diverse comprising many different classes of proteins such as transcription factors, hormone receptors, ubiquitin ligases and kinases, however they all require Hsp90 for proper folding and stabilization in order to gain functional activity [21]. We have clearly demonstrated that ANP32C and ANP32D show the characteristics of Hsp90 client proteins as their interaction is stabilized by molybdate ions and unperturbed under high salt washing conditions (Fig. 2c and Supplementary Fig. 6). The criterion of molybdate stabilization is sometimes not absolutely specific to clients. The interaction of Hsp90 with its co-chaperone Aha1 has been also shown to be strengthened by molybdate ions [34] however only molybdate-frozen Hsp90:client complexes seem to be resistant to high salt treatment [32–34]. The ANP32C and ANP32D proteins are not unique LRR-containing clients of Hsp90 since Leucine-Rich Repeat Kinase 2 (LRRK-2) [36], nucleotide-binding domain and leucine-rich repeat (NLR) containing proteins [37] and Scribble (Scrib) [38] have all been described as examples of LLR-containing proteins that require Hsp90 chaperoning activity. The ANP32C/D LRR-domain appears to be the region responsible for interaction with Hsp90 as ANP32D lacks the extended C-terminal region that is found in other ANP32 family members (see Sequence Alignment, Supplementary Fig. 9). At the same time the LRR-domain of other ANP32 family members despite high sequence similarity does not confer any association with Hsp90 since only the oncogenic ANP32C and ANP32D proteins were able to co-immunoprecipitate with Hsp90 in our experiments and ANP32A, B or E were not. This phenomenon of variable specificity of Hsp90 toward protein within the same family has been described for certain kinases, where it was concluded that thermal instability but not any particular structural motif is the major determinant of a Hsp90:kinase interaction [19]. In the same study a correlation between protein half-life time and the strength of interaction with Hsp90 was also described, where proteins having the shortest lifespan were shown to have the strongest interaction with Hsp90. This also appears consistent with our observations that other ANP32 family proteins remained inherently stable while ANP32C and ANP32D are quickly degraded when translation was blocked with cycloheximide. Indeed ANP32C/Y140H levels dropped to less than half of the initial

N

455 456

U

454

R O

O

F

Fig. 5. HeLa C and HeLa YH cells showed enhanced sensitivity to treatment with PU-H71: a) Wild type HeLa, HeLa C and HeLa YH were treated with indicated concentrations of PU-H71 for 30 h. Cell viability was measured by colorimetric CCK-8 assay. b) Wild type HeLa, HeLa C and HeLa YH cells were pre-treated with PU-H71 at 50 nM and 1 μM concentrations or DMSO (0.1%) as a vehicle control for 30 h and left for recovery and colony formation before fixation and staining with methylene blue.

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 Q5 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613

669

O

F

Conflicts of interest

R O

567 568

P

565 566

614 615

D

563 564

since the overexpression of ANP32A protein has been shown to cause the resistance of cells to the chemotherapeutic agent Gemcitabine while conveying sensitivity to 5-Fluoro Uracil [56]. We have also shown that ANP32C expression can convey enhanced resistance to the drug FTY720 [15]. PU-H71 has been shown to induce apoptosis in HeLa cells, as well as in melanoma, cervix, colon, liver and lung cancer cells, but not in normal human fibroblasts [57] implying that cells with high malignant potential like the hyperproliferative HeLa C and HeLa YH cells used in our study should display more sensitivity to drug effects in comparison to normal cells or in our case wild type HeLa cells. It is generally believed that Hsp90 inhibition causes cytostatic and/or cytotoxic effects via degradation of Hsp90 oncogenic clients. Aside from physical depletion Hsp90 dependent oncogenes can be inactivated via loss of function by Hsp90 inhibitors, as Hsp90 promotes not only folding and stability of its clients but also functional activation [58] and assists in multiprotein complex formation [59]. In case of HeLa C and HeLa YH cells, reduced viability in response to treatment with PU-H71 in comparison to wild type HeLa cells might be explained by a number of events with a cumulative cytostatic/cytotoxic effect. Firstly, ANP32C/ Y140H folding is compromised under conditions of Hsp90 inhibition. Secondly, it is plausible that Hsp90 inhibition could additionally lead to changes in modification/activity status of mature protein or their ability to participate in protein assemblies leading to a subsequent loss of function. Thirdly, since Hsp90 also plays a role in ANP32C/Y140H degradation either directly or indirectly, its inhibitor PU-H71 in the context of abrogated folding might lead to accumulation of the misfolded protein and formation of toxic protein aggregates. In case increased ANP32C/Y140H retains some functional potential, communication with downstream effectors might be prevented by PU-H71 treatment since it eliminates multiple drivers of cell proliferation, invasive potential and pro-survival proteins that have been described for PU-H71 in triple-negative breast cancer [60]. Which mechanisms are crucial for the PU-H71 cytotoxic effect on HeLa C/YH cells, or whether they all take place remains to be identified. Interestingly Hsp90 inhibition is of particular importance for prostate cancer since Hsp90 is overexpressed in prostate cancer cells compared with normal prostate epithelium [61] and it regulates the maturation and stability of many proteins that drive the development and progression of prostate cancer, including the androgen receptor [62]. The theory that Hsp90 inhibitors should show increased efficacy in cancers that are driven by Hsp90 client proteins, has also been supported by a number of clinical trials in breast cancer [63] and nonsmall cell lung carcinoma [64]. However, despite expression of ANP32C and ANP32D in tumors of the breast [8] and prostate [7] as well as reported expression in oral carcinoma cells [9] it is difficult to determine the clinical implications of our preliminary observations on cells treated with PU-H71 as details regarding the frequency of ANP32C/D overexpressing tumors are not available. The diverse molecular functions of ANP32 family proteins despite such close sequence identity remains a mysterious feature that requires further investigation and it remains to be seen what true role ANP32C and ANP32D might play in normal and malignant cells. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbamcr.2015.06.007.

E

561 562

T

559 560

C

557 558

E

555 556

R

554

R

552 553

N C O

550 551

to control (Supplementary Fig. 7), which could also be shown to immunoprecipitate significantly more Hsp90 from PU-H71 treated samples than the control (Fig. 4e). A small fraction (about 1%) of Hsp90 inhibition responsive proteome illustrates the same (downregulation followed by upregulation over the basal level) time-dependent pattern of quantitative changes. These results recently described in a T-cell proteome dynamics study upon geldanamycin treatment [40] suggest that ANP32C and its mutant are not unique in their behavior. Such a complex outcome of Hsp90 inhibition might be interpreted as a disturbance of two independent processes occurring simultaneously in the cell, one responsible for ANP32C stability and the other one performing ANP32C degradation under normal physiological conditions thus maintaining balanced steady state level of the protein. It's important to note that the level of ANP32C_Y140H protein ectopically expressed in our cellular models HeLa YH and ACHN YH is reasonably higher than the level of wild type ANP32C in HeLa C and ACHN C. The reason for that could be a shift in balance between degradation and stabilization toward the latter as seen in MG132 or PU-H71/17-AAG- treatment experiments (Supplementary Figs. 5 and 7). Therefore the more stable and abundant mutant ANP32C_Y140H is able to exert a stronger phenotype in cells [10,14]. The increased amount of ANP32C protein after prolonged inhibition of Hsp90 might be explained by Hsp90's direct action or through indirect effects such as degradation or loss of function of an as yet unidentified negative regulator. Indeed this has previously been observed in the case of p53 where a drop in MDM2 and MDMX proteins is accompanied by increased levels of p53 upon 17-AAG administration [41,42]. Another indirect effect of 17-AAG on protein abundance regardless of its client–non-client status is the increased transcription activity of the HSF1 transcription factor. This has been shown to be the cause of protein upregulation for several proteins including: synapsin I, synaptophysin, and PSD95 in neurons [43], IL-10 produced by Tregs of mice colon mucosa [44] and chaperones with co-chaperones. Differential protein regulation is a common response to the administration of geldanamycin or its derivatives 17-AGG and 17-DMAG. Although proteomic studies have revealed that protein down-regulation is most frequently observed, a significant portion of the proteome comprised mainly of proteins involved in: protein folding, unfolded protein response, proteasomal degradation, cytoskeletal proteins and metabolic enzymes is consistently up-regulated upon treatment with these drugs [45–47]. Additionally the oncogenic signaling pathways of Hsp90 inhibitor treated cells have also been recently studied using more modern stable isotope labeling by amino acids in cell culture (SILAC) techniques and revealed a large number of oncogenes enriched after 20 h of treatment with geldanamycin [40,48]. The most straightforward explanation for the increased level of ANP32C in response to treatment with 17-AAG/PU-H71 suggests a role for Hsp90 in its degradation. Indeed recent studies have shown that Hsp90-dependent folding and degradation are executed by distinct molecular complexes containing variable levels of Hsp70, CHIP and HOP activity that make folding predominant under normal conditions [49]. Hsp90 also actively participates in the proteasomal degradation of selected clients [35], including membrane proteins CYP2E1 [50], delta F508 CFTR mutant [51], apoprotein B [52], mutated forms of lysosomal enzyme glucocerebrosidase [53], misfolded or mutated cytosolic protein VHL [54], and chaperone-mediated autophagy for selective degradation of cytosolic proteins in lysosomes [55]. Interestingly, there are some reports of Hsp90 clients that do not undergo degradation when Hsp90 is inhibited with geldanamycin or PU-H71 but demonstrate subtle changes in their phosphorylation status (e.g. STAT3/5) [26,47]. If ANP32C changes its activity/modification pattern under conditions of Hsp90 inhibition remains to be determined. The treatment of ANP32C or ANP32C_Y140H overexpressing cells also conferred sensitivity to PU-H71, as demonstrated by CCK-8 proliferation and colony formation assays (Fig. 5). Modulation of drug resistance appears to be a common effect of ANP32 protein expression

U

548 549

9

The authors declare no financial or commercial conflict of interest.

616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668

670

Acknowledgments

671

YY is a graduate student of the HBRS “Molecular medicine” PhD program, Hannover Medical School. We thank Wiebke Hiemisch for her technical assistance, Dr Stephanie Weiss (Toplab GmBH, Martinsfreid, Germany) for carrying out the peptide mass fingerprinting

672

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

673 674 675

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

F

[1] A. Matilla, M. Radrizzani, The Anp32 family of proteins containing leucine-rich repeats, Cerebellum 4 (2005) 7–18. [2] P.T. Reilly, Y. Yu, A. Hamiche, L. Wang, Cracking the ANP32 whips: important functions, unequal requirement, and hints at disease implications, Bioessays (2014). [3] B. Kobe, A.V. Kajava, The leucine-rich repeat as a protein recognition motif, Curr. Opin. Struct. Biol. 11 (2001) 725–732. [4] T. Huyton, C. Wolberger, The crystal structure of the tumor suppressor protein pp 32 (Anp32a): structural insights into Anp32 family of proteins, Protein Sci. 16 (2007) 1308–1315. [5] J.R. Brody, S.S. Kadkol, M.A. Mahmoud, J.M. Rebel, G.R. Pasternack, Identification of sequences required for inhibition of oncogene-mediated transformation by pp 32, J. Biol. Chem. 274 (1999) 20053–20055. [6] T.H. Chen, J.R. Brody, F.E. Romantsev, J.G. Yu, A.E. Kayler, et al., Structure of pp 32, an acidic nuclear protein which inhibits oncogene-induced formation of transformed foci, Mol. Biol. Cell 7 (1996) 2045–2056. [7] S.S. Kadkol, J.R. Brody, J. Pevsner, J. Bai, G.R. Pasternack, Modulation of oncogenic potential by alternative gene use in human prostate cancer, Nat. Med. 5 (1999) 275–279. [8] S.S. Kadkol, G.A. El Naga, J.R. Brody, J. Bai, Y. Gusev, et al., Expression of pp 32 gene family members in breast cancer, Breast Cancer Res. Treat. 68 (2001) 65–73. [9] K. Imamachi, F. Higashino, T. Kitamura, W. Kakuguchi, A. Yanagawa-Matsuda, et al., pp32r1 controls the decay of the RNA-binding protein HuR, Oncol. Rep. 31 (2014) 1103–1108. [10] G.J. Kochevar, J.R. Brody, S.S. Kadkol, K.M. Murphy, G.R. Pasternack, Identification of a functional mutation in pp32r1 (ANP32C), Hum. Mutat. 23 (2004) 546–551. [11] J. Kochevar, H. Kessler, Mutant pp32r1 in oral carcinoma, Proc. Am. Assoc. Cancer Res. 47 (2006) (Abstract #1956). [12] D.H. Larsen, C. Poinsignon, T. Gudjonsson, C. Dinant, M.R. Payne, et al., The chromatin-remodeling factor CHD4 coordinates signaling and repair after DNA damage, J. Cell Biol. 190 (2003) 731–740. [13] Polo, S.E.; Kaidi, A.; Baskcomb, L.; Galanty, Y.; Jackson, S.P., Regulation of DNAdamage responses and cell-cycle progression by the chromatin remodelling factor CHD4. EMBO J 29: 3130–3139. [14] S. Buddaseth, W. Gottmann, R. Blasczyk, T. Huyton, Dysregulation of cell cycle control caused by overexpression of the oncogene pp32r1 (ANP32C) and the TyrNHis mutant pp32r1Y140H, Biochim. Biophys. Acta 1833 (2013) 1212–1221. [15] S. Buddaseth, W. Gottmann, R. Blasczyk, T. Huyton, Overexpression of the pp32r1 (ANP32C) oncogene or its functional mutant pp32r1Y140H confers enhanced resistance to FTY720 (Finguimod), Cancer Biol. Ther. 15 (2014) 289–296. [16] A. Rohl, J. Rohrberg, J. Buchner, The chaperone Hsp90: changing partners for demanding clients, Trends Biochem. Sci. 38 (2013) 253–262. [17] G. Chiosis, C.A. Dickey, J.L. Johnson, A global view of Hsp90 functions, Nat. Struct. Mol. Biol. 20 (2013) 1–4. [18] P.C. Echeverria, A. Bernthaler, P. Dupuis, B. Mayer, D. Picard, An interaction network predicted from public data as a discovery tool: application to the Hsp90 molecular chaperone machine, PLoS One 6 (2011) e26044. [19] M. Taipale, I. Krykbaeva, M. Koeva, C. Kayatekin, K.D. Westover, et al., Quantitative analysis of HSP90-client interactions reveals principles of substrate recognition, Cell 150 (2012) 987–1001. [20] A. Ahsan, S.G. Ramanand, C. Whitehead, S.M. Hiniker, A. Rehemtulla, et al., Wildtype EGFR is stabilized by direct interaction with HSP90 in cancer cells and tumors, Neoplasia 14 (2012) 670–677. [21] L. Whitesell, S.L. Lindquist, HSP90 and the chaperoning of cancer, Nat. Rev. Cancer 5 (2005) 761–772. [22] M.M. McCarthy, E. Pick, Y. Kluger, B. Gould-Rothberg, R. Lazova, et al., HSP90 as a marker of progression in melanoma, Ann. Oncol. 19 (2008) 590–594. [23] E. Pick, Y. Kluger, J.M. Giltnane, C. Moeder, R.L. Camp, et al., High HSP90 expression is associated with decreased survival in breast cancer, Cancer Res. 67 (2007) 2932–2937. [24] A. Kamal, F.J. Burrows, Hsp90 inhibitors as selective anticancer drugs, Discov. Med. 4 (2004) 277–280. [25] M.C. Diehl, M.O. Idowu, K. Kimmelshue, T.P. York, L.W. Elmore, et al., Elevated expression of nuclear Hsp90 in invasive breast tumors, Cancer Biol. Ther. 8 (2009) 1952–1961. [26] K. Moulick, J.H. Ahn, H. Zong, A. Rodina, L. Cerchietti, et al., Affinity-based proteomics reveal cancer-specific networks coordinated by Hsp90, Nat. Chem. Biol. 7 (2011) 818–826. [27] T. Taldone, A. Rodina, E.M. Dagama Gomes, M. Riolo, H.J. Patel, et al., Synthesis and evaluation of cell-permeable biotinylated PU-H71 derivatives as tumor Hsp90 probes, Beilstein J. Org. Chem. 9 (2013) 544–556. [28] K. Jhaveri, T. Taldone, S. Modi, G. Chiosis, Advances in the clinical development of heat shock protein 90 (Hsp90) inhibitors in cancers, Biochim. Biophys. Acta 1823 (2012) 742–755. [29] L. Neckers, P. Workman, Hsp90 molecular chaperone inhibitors: are we there yet? Clin. Cancer Res. 18 (2012) 64–76. [30] A.L. Smith, D.B. Friedman, H. Yu, R.H. Carnahan, A.B. Reynolds, ReCLIP (reversible cross-link immuno-precipitation): an efficient method for interrogation of labile protein complexes, PLoS One 6 (2011) e16206.

O

679 680 681 682 Q7 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 Q8 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756

R O

References

P

678

[31] T. Taldone, D. Zatorska, P.D. Patel, H. Zong, A. Rodina, et al., Design, synthesis, and evaluation of small molecule Hsp90 probes, Bioorg. Med. Chem. 19 (2011) 2603–2614. [32] S.D. Hartson, V. Thulasiraman, W. Huang, L. Whitesell, R.L. Matts, Molybdate inhibits hsp90, induces structural changes in its C-terminal domain, and alters its interactions with substrates, Biochemistry 38 (1999) 3837–3849. [33] B.T. Scroggins, T. Prince, J. Shao, S. Uma, W. Huang, et al., High affinity binding of Hsp90 is triggered by multiple discrete segments of its kinase clients, Biochemistry 42 (2003) 12550–12561. [34] L. Sun, T. Prince, J.R. Manjarrez, B.T. Scroggins, R.L. Matts, Characterization of the interaction of Aha1 with components of the Hsp90 chaperone machine and client proteins, Biochim. Biophys. Acta 1823 (2012) 1092–1101. [35] M.A. Theodoraki, A.J. Caplan, Quality control and fate determination of Hsp90 client proteins, Biochim. Biophys. Acta 1823 (2012) 683–688. [36] L. Wang, C. Xie, E. Greggio, L. Parisiadou, H. Shim, et al., The chaperone activity of heat shock protein 90 is critical for maintaining the stability of leucine-rich repeat kinase 2, J. Neurosci. 28 (2008) 3384–3391. [37] Y. Kadota, K. Shirasu, R. Guerois, NLR sensors meet at the SGT1-HSP90 crossroad, Trends Biochem. Sci. 35 (2010) 199–207. [38] D.J. Eastburn, M.M. Zegers, K.E. Mostov, Scrib regulates HGF-mediated epithelial morphogenesis and is stabilized by Sgt1-HSP90, J. Cell Sci. 125 (2012) 4147–4157. [39] T. Oda, T. Hayano, H. Miyaso, N. Takahashi, T. Yamashita, Hsp90 regulates the Fanconi anemia DNA damage response pathway, Blood 109 (2007) 5016–5026. [40] I. Fierro-Monti, P. Echeverria, J. Racle, C. Hernandez, D. Picard, et al., Dynamic impacts of the inhibition of the molecular chaperone Hsp90 on the T-cell proteome have implications for anti-cancer therapy, PLoS One 8 (2013) e80425. [41] J.L. Roh, E.H. Kim, H.B. Park, J.Y. Park, The Hsp90 inhibitor 17-(allylamino)-17demethoxygeldanamycin increases cisplatin antitumor activity by inducing p53mediated apoptosis in head and neck cancer, Cell Death Dis. 4 (2013) e956. [42] A.V. Vaseva, A.R. Yallowitz, N.D. Marchenko, S. Xu, U.M. Moll, Blockade of Hsp90 by 17AAG antagonizes MDMX and synergizes with Nutlin to induce p53-mediated apoptosis in solid tumors, Cell Death Dis. 2 (2011) e156. [43] Y. Chen, B. Wang, D. Liu, J.J. Li, Y. Xue, et al., Hsp90 chaperone inhibitor 17-AAG attenuates Abeta-induced synaptic toxicity and memory impairment, J. Neurosci. 34 (2014) 2464–2470. [44] C.B. Collins, C.M. Aherne, A. Yeckes, K. Pound, H.K. Eltzschig, et al., Inhibition of N-terminal ATPase on HSP90 attenuates colitis through enhanced Treg function, Mucosal Immunol. 6 (2013) 960–971. [45] A. Maloney, P.A. Clarke, S. Naaby-Hansen, R. Stein, J.O. Koopman, et al., Gene and protein expression profiling of human ovarian cancer cells treated with the heat shock protein 90 inhibitor 17-allylamino-17-demethoxygeldanamycin, Cancer Res. 67 (2007) 3239–3253. [46] K. Sharma, R.M. Vabulas, B. Macek, S. Pinkert, J. Cox, et al., Quantitative proteomics reveals that Hsp90 inhibition preferentially targets kinases and the DNA damage response, Mol. Cell. Proteomics 11 (M111) (2012) 014654. [47] Z. Wu, A. Moghaddas Gholami, B. Kuster, Systematic identification of the HSP90 candidate regulated proteome, Mol. Cell. Proteomics 11 (M111) (2012) 016675. [48] I. Fierro-Monti, J. Racle, C. Hernandez, P. Waridel, V. Hatzimanikatis, et al., A novel pulse-chase SILAC strategy measures changes in protein decay and synthesis rates induced by perturbation of proteostasis with an Hsp90 inhibitor, PLoS One 8 (2013) e80423. [49] L. Kundrat, L. Regan, Balance between folding and degradation for Hsp90-dependent client proteins: a key role for CHIP, Biochemistry 49 (2010) 7428–7438. [50] T. Goasduff, A.I. Cederbaum, CYP2E1 degradation by in vitro reconstituted systems: role of the molecular chaperone hsp90, Arch. Biochem. Biophys. 379 (2000) 321–330. [51] W. Fuller, A.W. Cuthbert, Post-translational disruption of the delta F508 cystic fibrosis transmembrane conductance regulator (CFTR)-molecular chaperone complex with geldanamycin stabilizes delta F508 CFTR in the rabbit reticulocyte lysate, J. Biol. Chem. 275 (2000) 37462–37468. [52] V. Gusarova, A.J. Caplan, J.L. Brodsky, E.A. Fisher, Apoprotein B degradation is promoted by the molecular chaperones hsp90 and hsp70, J. Biol. Chem. 276 (2001) 24891–24900. [53] J. Lu, C. Yang, M. Chen, D.Y. Ye, R.R. Lonser, et al., Histone deacetylase inhibitors prevent the degradation and restore the activity of glucocerebrosidase in Gaucher disease, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 21200–21205. [54] A.J. McClellan, M.D. Scott, J. Frydman, Folding and quality control of the VHL tumor suppressor proceed through distinct chaperone pathways, Cell 121 (2005) 739–748. [55] A.M. Cuervo, E. Wong, Chaperone-mediated autophagy: roles in disease and aging, Cell Res. 24 (2014) 92–104. [56] T.K. Williams, C.L. Costantino, N.A. Bildzukewicz, N.G. Richards, D.W. Rittenhouse, et al., pp32 (ANP32A) expression inhibits pancreatic cancer cell growth and induces gemcitabine resistance by disrupting HuR binding to mRNAs, PLoS One 5 (2010) e15455. [57] C. Gallerne, A. Prola, C. Lemaire, Hsp90 inhibition by PU-H71 induces apoptosis through endoplasmic reticulum stress and mitochondrial pathway in cancer cells and overcomes the resistance conferred by Bcl-2, Biochim. Biophys. Acta 1833 (2013) 1356–1366. [58] C.K. Vaughan, M. Mollapour, J.R. Smith, A. Truman, B. Hu, et al., Hsp90-dependent activation of protein kinases is regulated by chaperone-targeted dephosphorylation of Cdc37, Mol. Cell 31 (2008) 886–895. [59] T. Makhnevych, W.A. Houry, The role of Hsp90 in protein complex assembly, Biochim. Biophys. Acta 1823 (2012) 674–682. [60] E. Caldas-Lopes, L. Cerchietti, J.H. Ahn, C.C. Clement, A.I. Robles, et al., Hsp90 inhibitor PU-H71, a multimodal inhibitor of malignancy, induces complete responses in

D

and Professor Gabriella Chiosis (Memorial Sloan-Kettering Cancer Centre, New York, USA) for providing PU-H71 coupled beads.

U

N

C

O

R

R

E

C

T

676 677

E

10

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007

757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842

Y. Yuzefovych et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

[63] S. Modi, A. Stopeck, H. Linden, D. Solit, S. Chandarlapaty, et al., HSP90 inhibition is 850 effective in breast cancer: a phase II trial of tanespimycin (17-AAG) plus 851 trastuzumab in patients with HER2-positive metastatic breast cancer progressing 852 on trastuzumab, Clin. Cancer Res. 17 (2011) 5132–5139. [64] L.V. Sequist, S. Gettinger, N.N. Senzer, R.G. Martins, P.A. Janne, et al., Activity of IPI-504, a novel heat-shock protein 90 inhibitor, in patients with molecularly defined nonsmall-cell lung cancer, J. Clin. Oncol. 28 (2010) 4953–4960.

N C O

R

R

E

C

T

E

D

P

R O

O

F

triple-negative breast cancer models, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 8368–8373. [61] M.R. Cardillo, F. Ippoliti, IL-6, IL-10 and HSP-90 expression in tissue microarrays from human prostate cancer assessed by computer-assisted image analysis, Anticancer Res. 26 (2006) 3409–3416. [62] M.M. Centenera, A.K. Fitzpatrick, W.D. Tilley, L.M. Butler, Hsp90: still a viable target in prostate cancer, Biochim. Biophys. Acta 1835 (2013) 211–218.

U

843 844 845 846 847 848 849

11

Please cite this article as: Y. Yuzefovych, et al., Oncogenic acidic nuclear phosphoproteins ANP32C/D are novel clients of heat shock protein 90, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbamcr.2015.06.007