Kinase Subs Protein Array

Clinica Chimica Acta 357 (2005) 180 – 183 www.elsevier.com/locate/clinchim

Kinase substrate protein microarray analysis of human colon cancer and hepatic metastasis Claudio Bellucoa,*, Enzo Mammanoa, Emanuel Petricoinb, Luca Prevedelloa, Valerie Calvertb, Lance Liottac, Donato Nittia, Mario Lisea a Department of Oncological and Surgical Science, Surgery Branch, University of Padova, Padova, Italy FDA-NCI Clinical Proteomics Program, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, MD, USA c FDA-NCI Clinical Proteomics Program, Laboratory of Pathology, National Cancer Institute, Bethesda, MD, USA

b

Received 4 March 2005; accepted 9 March 2005 Available online 25 May 2005

Abstract Background: Liver metastases represent the major determinant of survival in patients with colorectal cancer (CRC). In cases with unresectable liver disease, more effective agents are needed, since chemotherapy achieves median survival of only 15 months. Protein kinases coordinate complex functions that are often disregulated in cancer and are therefore considered important targets for molecular therapeutics. In this study, we investigated the phosphoproteomic status of different protein kinases in primary CRC and in liver metastases. Methods: The status of 29 key endpoints was evaluated using reverse phase protein array on laser capture microdissected neoplastic cells from five primary CRCs without metastases, three patient-matched primary CRCs and synchronous liver metastases and five CRC metachronous liver metastases. Results: Unsupervised hierarchical two-way clustering analysis showed an entirely different phosphoproteomic profile in primary CRCs compared to liver metastases. This difference was observed also in primary and metastatic patient-matched lesions. Conclusions: Our findings of different signaling pathways between primary and metastatic CRC suggest a possible microenvironment effect, and emphasize the need to perform molecular network analysis of metastatic tissue when molecular targeting is considered. D 2005 Elsevier B.V. All rights reserved. Keywords: Proteomics; Colorectal cancer; Metastasis; Protein kinases

1. Introduction * Corresponding author. Tel.: +39 049 8211237/8212055; fax: +39 049 651891. E-mail address: [email protected] (C. Belluco). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cccn.2005.03.024

Colorectal cancer (CRC) is the second leading cause of cancer-related death in western countries and the liver represents the most common site of

C. Belluco et al. / Clinica Chimica Acta 357 (2005) 180–183

metastatic disease and the major determinant of survival in patients with this type of neoplasm [1]. About 70% of patients with CRC liver metastases have unresectable disease and, in these patients, chemotherapy achieves median survival of only 15 months [2]. New and more effective agents are therefore needed for the treatment of these patients. Extracellular stimuli, such as hormones, growth factors and cytokines, regulate biological cell functions by altering the levels of protein phosphorylation, which represent the initial and crucial event for most signaling pathways inside the cell. Protein kinases, by phosphorylating substrate proteins, direct the activity, localization and overall function of proteins involved in complex functions such as cell growth and differentiation, cell cycle control and apoptosis, which are often disregulated in cancer development and progression [3,4]. Protein kinases are, therefore, considered important targets for molecular therapeutics [5]. Reverse phase protein microarray is a new technique, which has been recently developed to map the state of key signal transduction pathways from human biopsy specimens by looking at dozens of kinase substrates at once through multiplexed phospho-specific antibody analysis [6]. In this study, we investigated by reverse phase protein microarray the phosphoproteomic status of different protein kinases in primary CRC and in liver metastases.

2. Materials and methods Frozen tissue samples from five primary CRCs without distant metastases, three patient-matched primary CRCs and synchronous liver metastases and five CRC metachronous liver metastases were used for this study. Tissue samples were collected directly in the operating room and immediately snap frozen and stored in liquid nitrogen until used. Eight-micron sections were obtained and approximately 20,000 cells from each tissue sample were microdissected by Laser Capture System (Arcturus Engineering, Mountain View, CA, USA). Cells were lysed for 30 min at 75 8C using a 1:1 mixture TPER Reagent (Pierce, Rockford, IL, USA) and 2 Tris–Glycine SDS Sample Buffer (Novex/Invitrogen). After cell lysis, cells were boiled for 10 min and stored at

181

4 8C for arraying. Three nanoliters of lysate were arrayed by a pin and ring GMSE 417 Arrayer (Affymetrix, Santa Clara, CA, USA) onto nitrocellulose slides. Arrayed slides were prepared for staining by reacting with Reblot (Chemicon, Temencula, CA, USA) followed by two washes with PBS washing buffer. Slides were then treated overnight with IBlock (Applied Biosystem, Bedford, MA, USA). To estimate the total protein amount, selected arrays were stained with Sypro Ruby Protein Blot Stain (Molecula Probes, Eugene, OR, USA) and visualized on a Fluorchemk imagin system (Alpha Innotech, San Leandro, CA, USA). Slides were stained on an automated slide stainer (Dako, Carpinteria, CA, USA) using a biotin-linked peroxidase catalyzed signal amplification. Finally, the primary antibodies at concentrations ranging from 1:50 to 1:1000 were applied for 30 min followed by the secondary link antibody for 30 min (concentration 1:10 for antimouse antibodies and 1:5000 for antirabbit antibodies). Twenty nine commercially available primary antibodies were used: Cleaved Caspase 3 (D175) 1:50 (Cell Signaling Technology, Beverly, MA, USA), Phospho-CREB (S133) 1:200 (Cell Signaling Technology), Phospho-ERK 1/2 (T202/Y204) 1:500 (Cell Signaling Technology), Phospho-IRS1 (S616) 1:500 (Biosource, Camarillo, CA, USA), Phospho-IKBa (S32) 1:100 (Cell Signaling Technology), EGFR 1:50 (Cell Signaling Technology), pGSK-3 alpha/beta (Ser21/9) 1:1000 (Biosource), Phospho-cKit (Y719) 1:100 (Cell Signaling Technology), Phospho-EGFR (Y1148) 1:100 (Biosource), Phospho-STAT1 (Y701) 1:200 (Cell Signaling Technology), ErbB2 1:50 (Neomarkers, Lab Vision Corporation, Fremont, CA, USA), PhosphoPDGFRb (Y716) 1:100 (Upstate, Waltham, MA, USA), Phospho-PKCa´ (S657) 1:1000 (Upstate), Cox2 1:200 (Upstate), Phospho-mTOR (S2448) 1:100 (Cell Signaling Technology), PhosphoMARCKS (S152/156) 1:50 (Cell Signaling Technology), Phospho-P38 (T180/Y182) 1:50 (Cell Signaling Technology), Phospho-PTEN (S380) 1:200 (Cell Signaling Technology), Phospho-STAT3 (S727) 1:50 (Biosource), Phospho-Aurora A/AIK (T288) 1:100 (Cell Signaling Technology), Phospho-Zap 70 (Y493) 1:200 (Cell Signaling Technology), pJNK/ SAPK (Thr183/Tyr185) 1:50 (Cell Signaling Technology), pNF-kB p65 (Ser536) 1:50 (Upstate), Cleaved Caspase 9 (D315) 1:50 (Cell Signaling Technology),

182

C. Belluco et al. / Clinica Chimica Acta 357 (2005) 180–183

Phospho-BAD (S112) 1:100 (Cell Signaling Technology), Phospho-c-Abl (T735) 1:50 (Cell Signaling Technology), Phospho-STAT1 (S727) 1:1000 (Upstate), Phospho-AKT (S473) 1:50 (Cell Signaling Technology), Phospho-ErbB2/Her2 (Y1248) 1:1000 (Upstate). The specificity of each antibody was previously tested by Western blotting. Stained slides were scanned individually on a UMAX PowerLook III scanner (UMAX, Dallas TX, USA) at 600 dpi and saved as TIF files. The TIf images for antibobystained and Sypro-stained slides images were analyzed with ImageQuant 5.2 (Molecular Dynamics, Sunyvale, CA, USA). For each antibody, the average

pixel intensity value for the first point in each dilution curve was divided by the corresponding value of the Sypro-stained total potein slide. Unsupervised hierarchical two-way clustering analysis was used for comparing the signaling profiles of the different samples.

3. Results Unsupervised hierarchical two-way clustering analysis of the signaling pathway using 29 different endpoints demonstrated that the phosphoproteomic profile of the liver metastases was entirely different

Fig. 1. Unsupervised hierarchical clustering analysis of primary and metastatic CRC using 29 phosphoproteomic endpoints. The phosphoproteomic profile of the liver metastases is entirely different from that of the primary tumors. (Colon Primary: primary CRC without distant metastasis; Colon Pri/Met: patient-matched primary CRC with synchronous liver metastasis; Metachr CRC: metachronous CRC liver metastasis; Liver Met: patient-matched synchronous liver metastasis).

C. Belluco et al. / Clinica Chimica Acta 357 (2005) 180–183

from that of the primary tumors. This difference was also observed in the three patient-matched primary CRCs and liver metastases. Overall, most of the protein kinases appeared to be overexpressed in the liver metastases compared to the primary tumors. A representative example of the array layout is shown in Fig. 1.

4. Conclusions While gene microarrays can provide important information about somatic genetic taxonomy, they are unable to provide a full picture of the fluctuating signaling events that occur at the proteomic level. In fact, cellular signaling events are driven by protein– protein interactions, post-translational protein modifications and enzymatic activities that cannot be accurately predicted or described by transcriptional profiling methods alone [4,7]. Recently, antibodies have been developed to specifically recognize the phosphorylated isoform of kinase substrates. In theory, it could be possible to evaluate the state of entire portions of a signaling pathway or cascade, even though the cell is lysed, by looking at dozens of kinase substrates at once through multiplexed phospho-specific antibody analysis. Protein microarrays offer the promise to dramatically multiplex, quantify, accelerate and miniaturize this type of analysis over any existing format [8,9]. Our study demonstrates that reverse phase protein microarray analysis of human primary CRC and liver metastases is feasible and requires only approximately 20,000 cancer cells to perform multiplexed phosphoproteomic fingerprinting of a large number of signaling endpoints simultaneously. This analysis can be performed using new software tools that employ well-founded statistical methods to compare relative levels of phospho-specific endpoints across different tissues from several different patients. Unsupervised hierarchical clustering analysis of the signaling pathway portraits identified new signaling circuitry that was associated with a metastatic specific moleculare network and a primary tumor specific circuit. Clustering analysis revealed that the signaling networks in

183

use within primary CRC dramatically change upon metastasis. This finding, which might be explained by a direct influence of the organ microenvironment on the neoplastic cells, re-emphasizes the need to perform molecular network analysis of the metastatic process itself when therapeutic targeting is considered. Moreover, analysis and identification of important metastatic signaling networks could yield new insights into target selection for targeted therapeutics. In the future, this new technique may be used to map the state of key pathways of patient’s tissue samples before starting chemotherapeutic treatment to choose an individualized and optimized combination therapy. Since our findings are of potential clinical relevance, further studies using larger study sets will need to confirm our data and to evaluate if the same type of tumor has different phosphorylated phenotypes in different metastatic organs.

References [1] Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, Ward E, et al. American Cancer Society Cancer Statistics, 2004. CA Cancer J Clin 2004;54(1):8 – 29. [2] Liu LX, Zhang WH, Jiang HC. Current treatment for liver metastases from colorectal cancer. World J Gastroenterol 2003; 9(2):193 – 200. [3] Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase complement of the human genome. Science 2002;298(5600):1912 – 34. [4] Blume-Jensen P, Hunter T. Oncogenic kinase signalling. Nature 2001;411(6835):355 – 65. [5] Petricoin EF, Liotta LA. Molecular profiling of human cancer. Nat Rev Genet 2000;1:48 – 56. [6] Paweletz CP, Charboneau L, Bichsel VE, Simone NL, Chen T, Gillespie JW, et al. Reverse phase protein microarrays which capture disease progression show activation of pro-survival pathways at the cancer invasion front. Oncogene 2001; 20(16):1981 – 9. [7] Ge H, Walhout AJ, Vidal M. Integrating domicT information: a bridge between genomics and systems biology. Trends Genet 2003;19:551 – 60. [8] Nielsen UB, Cardone MH, Sinskey AJ, MacBeath G, Sorger PK. Profiling receptor tyrosine kinase activation by using Ab microarrays. Proc Natl Acad Sci U S A 2003;100:9330 – 5. [9] Liotta LA, Espina V, Mehta AI, Calvert V, Rosenblatt K, Geho D, et al. Protein microarrays: meeting analytical challenges for clinical applications. Cancer Cell 2003;3(4):317 – 25.