Western blot analysis and immunoprecipitation
Protein expression analysis in the panel of seven established breast cancer cell lines was assessed by Western blotting. Total cell extracts were prepared and electrophoresed in SDS-polyacrylamide gel and transferred to polyvinylidene fluoride (PVDF) membrane (Millipore Corporation, Billerica, MA, USA). The following antibodies were used for immunoblotting: Her3 (sc-285; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), Her2 (ab8054; Abcam, Cambridge, UK), phosphor-Her2 (Tyr1248) (44�, Invitrogen Corporation, Carlsbad, CA, USA) EGFR (#2232; Cell Signaling Technology, Inc., Danvers, MA, USA), phosphor-Akt (Ser473) (#9271; Cell Signaling Technology, Inc.), Akt (#9272, Cell Signaling Technology, Inc.), ER-α (sc-542; Santa Cruz Biotechnology, Inc.), PI3K (p110α) (#4254; Cell Signaling Technology, Inc.), PDK1 (#3062; Cell Signaling Technology, Inc.), HSP70 (SPA-810; Stressgen Bioreagents, now part of Assay Designs, Inc., Ann Arbor, MI, USA), HSP90 (SPA-845; Assay Designs, Inc.), Hsc70 (sc-7298, Santa Cruz Biotechnology, Inc.), Rb (#9309; Cell Signaling Technology, Inc.), pMEK1/2 (#9121, Cell Signaling Technology, Inc.), pERK (#9101; Cell Signaling Technology, Inc.), Bax (#2772, Cell Signaling Technology, Inc.), Bcl-2 (sc-492; Santa Cruz Biotechnology, Inc.), Bad (#9292; Cell Signaling Technology, Inc.), Bad (#9292; Cell Signaling Technology, Inc.), and Bcl-XL (#2762; Cell Signaling Technology, Inc.).
To assess total levels of AKT, phosphorylated AKT, β-tubulin, and ERBB2 in the cell extracts, 20 to 30 μg of total protein was resolved by the appropriate-percentage SDS-PAGE. The following antibodies were used for immunoblotting: anti-AKT (cat. no. 9272, rabbit polyclonal; Cell Signaling Technology, Inc.), anti-phospho-AKT (Ser473) (cat. no. 9271, rabbit polyclonal; Cell Signaling Technology, Inc.), anti-β-tubulin (cat. no. T 4026, mouse monoclonal, clone Tub2.1; Sigma-Aldrich, St. Louis, MO, USA), and anti-ERBB2 (cat. no. 28-0004, rabbit polyclonal; Zymed Laboratories Inc., now part of Invitrogen Corporation, Carlsbad, CA, USA). Bound antibodies on PVDF immunoblots were detected by Amersham ECL (Amersham, now part of GE Healthcare, Little Chalfont, Buckinghamshire, UK) or infrared fluorescence detection (LI-COR Biosciences, Lincoln, NE, USA).
For immunoprecipitation, 300 μg of total protein was immunoprecipitated with 5 μg of rat anti-HSP90α monoclonal antibody (SPA-840, clone 9D2, isotype: IgG2a; Assay Designs, Inc.). Proteins were resolved by 12% SDS-PAGE and transferred to PVDF membranes. The antibodies used were an anti-HSP90α antibody (cat. no. SPS-771, rabbit polyclonal; Assay Designs, Inc.) and an anti-p23 antibody (cat. no. ALX-804-023, mouse monoclonal, clone JJ3, isotype: IgG1; ALEXIS Corporation, Lausen, Switzerland).
Immunohistochemistry
Formalin-fixed paraffin-embedded tissue sections were stained using the Ventana Discovery System (Ventana Medical Systems, Inc., Tucson, AZ, USA). Deparaffinization of tissue sections and heat-induced epitope retrieval using Standard Cell Conditioning Solution 1 (Ventana Medical Systems, Inc.) were performed directly on the System. A rabbit anti-human ERBB2 (HER2) monoclonal antibody (Lab Vision Corporation, Fremont, CA, USA) and a mouse anti-human HSP70 (Assay Designs, Inc.) were prepared in Dako diluent (Dako North America, Inc., Carpinteria, CA, USA) and used at concentrations of 2 ug/mL for 32 minutes and 10 ug/mL for 60 minutes, respectively. For HSP70 staining, slides were incubated with a biotin-labeled anti-mouse IgG1 (Research Diagnostics, Inc., now known as Fitzgerald Industries International, Concord, MA, USA) at a concentration of 1.25 ug/mL diluted in M.O.M. (Mouse-on-Mouse) (Vector Laboratories, Peterborough, UK). Detection using a 3,3'-diaminobenzadine reaction was performed on the section by using the Ventana OmniMAP DAB for the ERBB2 antibody and the Ventana DAB Map reagent for the HSP70 antibody (Ventana Medical Systems, Inc.). Each tissue section was subsequently counterstained with hematoxylin. To ensure antibody specificity, consecutive tissue sections were incubated with normal isotype-matched immunoglobulins (rabbit IgG; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA, and mouse IgG1; Lab Vision Corporation) used at concentrations equivalent to Her2 and HSP70 antibodies. Stained tissue sections were quantified using the Aperio Digital Pathology System (Aperio Technologies, Inc., Vista, CA, USA).
Breast cancer xenograft model and efficacy studies
The ER-positive ERBB2-overexpressing cell line BT-474, which initially was derived from a human breast ductal carcinoma established from a solid invasive ductal carcinoma of the breast of a 60-year-old woman, was purchased from the ATCC (HTB-20). The cells were grown in DMEM high glucose (4.5 g/L) supplemented with 10% FCS, 200 mM l-glutamine, and 1% sodium pyruvate (BioConcept, Allschwil, Switzerland). Two or three days prior to cell inoculation, each mouse was subcutaneously implanted on the upper dorsal side with a 17β-estradiol pellet (25 μg/day, 90-day release; Innovative Research of America, Sarasota, FL, USA) using a trocar needle. BT-474 cells (5 × 106) were injected in 200 μL of Matrigel/Hanks' balanced salt solution (1:1 vol) (BD Matrigel™ Basement Membrane Matrix; BD Biosciences, San Jose, CA, USA) subcutaneously in the right flank. Invasive procedures were performed under Forene anesthesia. All experiments were performed using female Harlan HsdNpa: Athymic Nude-nu mice that were obtained from Novartis internal breeding stocks (Laboratory Animal Services, Novartis Pharma AG, Basel, Switzerland). The animals were kept under optimized hygienic conditions with 12-hour dark/12-hour light conditions. The animals were fed food and water ad libitum. All animal experiments were performed in strict adherence to the Swiss law for animal protection. The experimental protocols were approved by the Swiss Cantonal Veterinary Office of Basel-Stadt.
Pharmacokinetic analysis
Female athymic BT-474 tumor-bearing mice with tumors of approximately 250 mm3 received an i.v. dose of 30 mg/kg of NVP-AUY922. At various time points, mice (n = 4) were sacrificed and blood and tissues (tumor, liver, lung, heart, and muscles) were dissected. Concentrations of NVP-AUY922 in plasma and tissues were determined by high-pressure liquid chromatography/tandem mass spectrometry (HPLC/MS-MS) operated in electrospray ionization-positive mode. Frozen tissues were minced, then homogenized in an equal volume of ice-cold phosphate-buffered saline (Sigma P4417; Sigma-Aldrich) using a Polytron homogenizer (TP18-10; IKA, Staufen, Germany) and keeping the material cold during the homogenization. After the addition of 50 μL of internal standard (1 μg/mL) to analytical aliquots (25 to 250 μL) of plasma or tissue homogenate, the proteins were precipitated by the addition of an equal volume of acetonitrile and processed further for chromatographic separation. After three repetitions of protein precipitation by the addition of an equal volume of acetonitrile followed always by evaporation to dryness, the samples were redissolved in 100 μL of acetonitrile/water (1/9 vol/vol) containing 0.2% vol/vol formic acid. An aliquot (5 μL) of this solution was separated on a RESECT™ Ultra Cyano reverse-phase HPLC column (column size 50 × 1 mm, particle size 3 μm, preceded by a guard column: Phenomenex™ AJO-4304 Phenylpropyl, size 4 × 2 mm (Phenomenex, Torrance, CA, USA)) with a mobile phase consisting of a mixture of 0.2% formic acid in water (solvent A) and 0.2% formic acid in acetonitrile (solvent B). The column eluent was introduced directly into the ion source of the triple-quadrupole mass spectrometer Quattro Ultima™ (Micromass Limited, now part of Waters Corporation, Milford, MA, USA) controlled by Masslynx™ 4.0 software. Positive electrospray ionization multiple-reaction monitoring was used for the MS/MS detection of the analyte. Precursors to product ion transitions of m/z 466.35 → m/z 308.20 for NVP-AUY922-NX and m/z 480.40 → m/z 308.15 for IS VER814 were used. The limits of quantification were set to 4 ng/mL and 10 ng/g for plasma and tissues, respectively (coefficient of variation and overall bias less than 30%). Regression analysis and further calculations were performed using QuanLynx™ 4.0 (Waters Corporation, Milford, MA, USA) and Excel™ 2002 (Microsoft Corporation, Redmond, WA, USA). Concentrations of unknown samples were calculated from the peak area ratio of the product ion of the analytes to the product ion of its internal standard (ordinate) against the nominal concentration (abscissa). Assay linearity was indicated by an overall regression coefficient of 0.9975.