Recent technical advances in digital image capture and analysis greatly improve the measurement of protein expression in tissues. Breast cancer biomarkers provide a unique opportunity to utilize digital image analysis to evaluate sources of variability that are caused by the tissue preparation, in particular the decalcification treatment associated with the analysis of bone metastatic breast cancer, and to develop methods for comparison of digital data and categorical scores rendered by pathologists.
Tissues were prospectively decalcified for up to 24 hours and stained by immunohistochemistry (IHC) for ER, PR, Ki-67 and p53. HER2 positive breast cancer sections were retrieved from the pathology archives, and annotated with the categorical HER2 expression scores from the pathology reports. Digital images were captured with Leica and Aperio slide scanners. The conversion of the digital to categorical scores was accomplished with a Gaussian mixture model and tested for accuracy by comparison to clinical scores.
We observe significant effects of the decalcification treatment on common breast cancer biomarkers that are used in the clinic. ER, PR and p53 staining intensities decreased 15 – 20%, whereas Ki-67 decreased > 90% during the first 6 hrs of treatment and stabilized thereafter. In comparison with the Aperio images, pixel intensities generated by the Leica system are lower. A novel statistical model for conversion of digital to categorical scores provides a systematic approach for conversion of nuclear and membrane stains and demonstrated a high concordance with clinical scores.
Digital image analysis greatly improves the quantification of protein expression in human tissues. Decalcification affects the accuracy of immunohistochemical staining results and cannot be reversed by image analysis. Measurement data obtained on a continuous scoring scale can be converted to categorical scores for comparison with categorical dataset that are generated by pathologists.
The virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/13000_2014_213
Electronic supplementary material
The online version of this article (doi:10.1186/s13000-014-0213-9) contains supplementary material, which is available to authorized users.
Tissue decalcification; Breast cancer biomarkers; Image analysis; Quantification
Prostate cancer (PCa) remains a principal cause of mortality in developed countries. Because no clinical interventions overcome resistance to androgen ablation therapy, management of castration resistance and metastatic disease remains largely untreatable. Metastasis is a multistep process in which tumor cells lose cell-cell contacts, egress from the primary tumor, intravasate, survive shear stress within the vasculature and extravasate into tissues to colonize ectopic sites. Tumor cells reestablish migratory behaviors employed during nonneoplastic processes such as embryonic development, leukocyte trafficking and wound healing. While mesenchymal motility is an established paradigm of dissemination, an alternate, ‘amoeboid’ phenotype is increasingly appreciated as relevant to human cancer. Here we discuss characteristics and pathways underlying the phenotype, and highlight our findings that the cytoskeletal regulator DIAPH3 governs the mesenchymal-amoeboid transition. We also describe our identification of a new class of tumor-derived microvesicles, large oncosomes, produced by amoeboid cells and with potential clinical utility in prostate and other cancers.
The complexity of survival mechanisms in cancer cells from patients remains poorly understood. To obtain a comprehensive picture of tumor cell survival in prostate cancer metastases, we interrogated 5 survival proteins that operate within 3 survival pathways in a cohort of 185 lethal metastatic prostate metastases obtained from 44 patients. The expression levels of BCL-2, BCL-XL, MCL-1, cytoplasmic Survivin, nuclear Survivin, and Stathmin were measured by immunohistochemistry in a tissue microarray. Simultaneous expression of 3 or more proteins occured in 81% of lethal prostate cancer metastasis and BCL-2, cytoplasmic Survivin and MCL-1 were co-expressed in 71% of metastatic sites. An unsupervised cluster analysis separated bone and soft tissue metastases according to patterns of survival protein expression. BCL-2, cytoplasmic Survivin and MCL-1 had significantly higher expression in bone metastases (p < 10−5), while nuclear Survivin was significantly higher in soft tissue metastases (p = 3×10−14). BCL-XL overexpression in soft tissue metastases almost reached significance (p =0.09), while stathmin expression did not (p=0.28). In addition, the expression of MCL-1 was significantly higher in AR-positive tumors. Neuroendocrine differentiation was not associated with specific survival pathways. These studies demonstrate for the first time that bone and soft tissue metastases from the same patient differ significantly in expression of a panel of survival proteins and that with regards to survival protein expression, expression is associated with metastatic site and not patient. Altogether this study suggests that optimal therapeutic inhibition may require combinations of drugs that target both bone as well as soft tissue-specific survival pathways.
prostate cancer; metastases; anti-apoptotic pathways; tumor heterogeneity; immunohistochemistry
Tumor cells are inherently heterogeneous and often exhibit diminished adhesion, resulting in the shedding of tumor cells into the circulation to form circulating tumor cells (CTCs). A fraction of these are live CTCs with potential of metastatic colonization whereas others are at various stages of apoptosis making them likely to be less relevant to understanding the disease. Isolation and characterization of live CTCs may augment information yielded by standard enumeration to help physicians to more accurately establish diagnosis, choose therapy, monitor response, and provide prognosis. We previously reported on a group of near-infrared (NIR) heptamethine carbocyanine dyes that are specifically and actively transported into live cancer cells. In this study, this viable tumor cell-specific behavior was utilized to detect live CTCs in prostate cancer patients. Peripheral blood mononuclear cells (PBMCs) from 40 patients with localized prostate cancer together with 5 patients with metastatic disease were stained with IR-783, the prototype heptamethine cyanine dye. Stained cells were subjected to flow cytometric analysis to identify live (NIR+) CTCs from the pool of total CTCs, which were identified by EpCAM staining. In patients with localized tumor, live CTC counts corresponded with total CTC numbers. Higher live CTC counts were seen in patients with larger tumors and those with more aggressive pathologic features including positive margins and/or lymph node invasion. Even higher CTC numbers (live and total) were detected in patients with metastatic disease. Live CTC counts declined when patients were receiving effective treatments, and conversely the counts tended to rise at the time of disease progression. Our study demonstrates the feasibility of applying of this staining technique to identify live CTCs, creating an opportunity for further molecular interrogation of a more biologically relevant CTC population.
Two genome-wide association studies (GWAS) identified the β-microseminoprotein (MSMB) promoter SNP, rs10993994:C>T, as significantly associated with prostate cancer (PC) risk. Follow-up studies demonstrate that the variant allele directly affects expression of the MSMB encoded protein, PSP94, and also suggest that it affects mRNA expression levels of an adjacent gene, NCOA4, which is involved in androgen receptor transactivation. In a population-based study of 1,323 cases and 1,268 age-matched controls, we found the NCOA4 SNP, rs7350420:T>C, was associated with a 15% reduction in PC risk, but the association was not significant after adjustment for the rs10993994:C>T genotype. Tumor tissue microarrays of 519 radical prostatectomy patients were used to measure PSP94 and NCOA4 protein expression. Taken together, these data confirm that the rs10993994:C>T variant allele is associated with decreased PSP94 expression, and the association is stronger in tumor compared to normal prostate tissue. No association was observed between rs10993994:C>T and NCOA4 expression, and only moderate associations were seen between two NCOA4 SNPs, rs10761618:T>C and rs7085433:G>A, and NCOA4 protein expression. These data indicate that the increase in PC risk associated with rs10993994:C>T is likely mediated by the variant’s effect on PSP94 expression; however this effect does not extend to NCOA4 in the data presented here.
Prostate cancer; MSMB; PSP94; NCOA4; protein expression; genetic variants
MicroRNAs (miRNAs) are small (∼22 nucleotide) non-coding RNAs that regulate a myriad of biological processes and are frequently dysregulated in cancer. Cancer-associated microRNAs have been detected in serum and plasma and hold promise as minimally invasive cancer biomarkers, potentially for assessing disease characteristics in patients with metastatic disease that is difficult to biopsy. Here we used miRNA profiling to identify cancer-associated miRNAs that are differentially expressed in sera from patients with metastatic castration resistant prostate cancer (mCRPC) as compared to healthy controls. Of 365 miRNAs profiled, we identified five serum miRNAs (miR-141, miR-200a, miR-200c, miR-210 and miR-375) that were elevated in cases compared to controls across two independent cohorts. One of these, miR-210, is a known transcriptional target of the hypoxia-responsive HIF-1α signaling pathway. Exposure of cultured prostate cancer cells to hypoxia led to induction of miR-210 and its release into the extracellular environment. Moreover, we found that serum miR-210 levels varied widely amongst mCRPC patients undergoing therapy, and correlated with treatment response as assessed by change in PSA. Our results suggest that (i) cancer-associated hypoxia is a frequent, previously under-appreciated characteristic of mCRPC, and (ii) serum miR-210 may be further developed as a predictive biomarker in patients with this distinct disease biology.
Recent advances in tissue microarray technology have allowed immunohistochemistry to become a powerful medium-to-high throughput analysis tool, particularly for the validation of diagnostic and prognostic biomarkers. However, as study size grows, the manual evaluation of these assays becomes a prohibitive limitation; it vastly reduces throughput and greatly increases variability and expense. We propose an algorithm—Tissue Array Co-Occurrence Matrix Analysis (TACOMA)—for quantifying cellular phenotypes based on textural regularity summarized by local inter-pixel relationships. The algorithm can be easily trained for any staining pattern, is absent of sensitive tuning parameters and has the ability to report salient pixels in an image that contribute to its score. Pathologists’ input via informative training patches is an important aspect of the algorithm that allows the training for any specific marker or cell type. With co-training, the error rate of TACOMA can be reduced substantially for a very small training sample (e.g., with size 30). We give theoretical insights into the success of co-training via thinning of the feature set in a high dimensional setting when there is “sufficient” redundancy among the features. TACOMA is flexible, transparent and provides a scoring process that can be evaluated with clarity and confidence. In a study based on an estrogen receptor (ER) marker, we show that TACOMA is comparable to, or outperforms, pathologists’ performance in terms of accuracy and repeatability.
The primary objective of this study is to demonstrate the activation and analyze the regulation of the MEK- S6 kinase pathway in high-grade ovarian cancer. Phospho-ERK (pERK), a direct substrate of MEK and two phosphorylation sites on the ribosomal protein, S6, Ser235/236 and Ser240/244, which are both targeted by the MEK and PI3-kinase/AKT pathways, were analyzed in 13 cell lines, 28 primary cancers and 8 cases of cancer cells from ascites. In primary cancers, ERK and S6 phosphorylation was measured by immunohistochemistry (IHC). pERK, pS6, pAKT and p4EBP1 were also measured by Western blotting (WB). The regulation of S6 phosphorylation by the MEK and PI3-kinase pathways was determined in ovarian cancer cell lines. We observed frequent pERK expression in primary ovarian cancers (100 % by WB, 75% by IHC) but not in ovarian cancer cells from ascites (25% of cases by WB). The activation of the AKT pathway, measured by pAKT expression occurred in 7 cases of primary ovarian cancer by WB, but in none of the ascites samples. In ovarian cancer cell lines, the MEK pathway had a greater effect on S6 phosphorylation in cells without hyperactive AKT signaling. Our data suggest that MEK is a potential drug target in high-grade ovarian cancer, however cancer cells with hyperactive AKT and cancer cells in ascites may be less responsive to MEK inhibition. The phosphorylation of S6 as a specific biomarker for either MEK or PI3-kinase pathway activation should be used with caution.
ovarian cancer; pathway activation; MEK; AKT; S6
Several malignancies are known to exhibit a “field-effect” whereby regions beyond tumor boundaries harbor histological or molecular changes that are associated with cancer. We sought to determine if histologically benign prostate epithelium collected from men with prostate cancer exhibits features indicative of pre-malignancy or field effect.
Prostate needle biopsies from 15 men with high grade(Gleason 8–10) prostate cancer and 15 age- and BMI-matched controls were identified from a biospecimen repository. Benign epithelia from each patient were isolated by laser capture microdissection. RNA was isolated, amplified, and used for microarray hybridization. Quantitative PCR(qPCR) was used to determine the expression of specific genes of interest. Alterations in protein expression were analyzed through immunohistochemistry.
Overall patterns of gene expression in microdissected benign-associated benign epithelium (BABE) and cancer-associated benign epithelium (CABE) were similar. Two genes previously associated with prostate cancer, PSMA and SSTR1, were significantly upregulated in the CABE group(FDR <1%). Expression of other prostate cancer-associated genes, including ERG, HOXC4, HOXC5 and MME, were also increased in CABE by qRT-PCR, although other genes commonly altered in prostate cancer were not different between the BABE and CABE samples. The expression of MME and PSMA proteins on IHC coincided with their mRNA alterations.
Gene expression profiles between benign epithelia of patients with and without prostate cancer are very similar. However, these tissues exhibit differences in the expression levels of several genes previously associated with prostate cancer development or progression. These differences may comprise a field effect and represent early events in carcinogenesis.
Prostate cancer; gene regulation; carcinogenesis
The inappropriate expression of the c-MET cell surface receptor in many human solid tumors necessitates the development of companion diagnostics to identify those patients who could benefit from c-MET targeted therapies. Tumor tissues are formalin-fixed and paraffin embedded (FFPE) for histopathological evaluation, making the development of an antibody against c-MET that accurately and reproducibly detects the protein in FFPE samples an urgent need. We have developed a monoclonal antibody, designated MET4, from a panel of MET-avid monoclonal antibodies, based on its specific staining pattern in FFPE preparations of normal human prostate tissues. The accuracy of MET4 immunohistochemistry (MET4-IHC) was assessed by comparing MET4-IHC in FFPE cell pellets with immunoblotting analysis. The technical reproducibility of MET4-IHC possessed a percentage coefficient of variability (%CV) of 6.25% in intra-assay and inter-assay testing. Comparison with other commercial c-MET antibody detection reagents demonstrated equal specificity and increased sensitivity for c-MET detection in prostate tissues. In two cohorts of ovarian cancers and gliomas, MET4 reacted with ovarian cancers of all histological subtypes (strong staining in 25%) and with 63% of gliomas. In addition, MET4 bound c-Met on the surfaces of cultured human cancer cells and tumor xenografts. In summary, the MET4 monoclonal antibody accurately and reproducibly measures c-MET expression by IHC in FFPE tissues and can be used for molecular imaging in-vivo. These properties encourage further development of MET4 as a multipurpose molecular diagnostics reagent to help to guide appropriate selection of patients being considered for treatment with c-MET-antagonistic drugs.
c-MET; monoclonal antibody; ovarian cancer; glioma; molecular diagnostics; immunohistochemistry
Epithelial ovarian carcinoma is a significant cause of cancer mortality in women worldwide and in the United States. Epithelial ovarian cancer comprises several histological subtypes, each with distinct clinical and molecular characteristics. The natural history of this heterogeneous disease, including the cell types of origin, is poorly understood. This study applied recently developed methods for high-throughput DNA methylation profiling to characterize ovarian cancer cell lines and tumors, including representatives of three major histologies.
We obtained DNA methylation profiles of 1,505 CpG sites (808 genes) in 27 primary epithelial ovarian tumors and 15 ovarian cancer cell lines. We found that the DNA methylation profiles of ovarian cancer cell lines were markedly different from those of primary ovarian tumors. Aggregate DNA methylation levels of the assayed CpG sites tended to be higher in ovarian cancer cell lines relative to ovarian tumors. Within the primary tumors, those of the same histological type were more alike in their methylation profiles than those of different subtypes. Supervised analyses identified 90 CpG sites (68 genes) that exhibited ‘subtype-specific’ DNA methylation patterns (FDR<1%) among the tumors. In ovarian cancer cell lines, we estimated that for at least 27% of analyzed autosomal CpG sites, increases in methylation were accompanied by decreases in transcription of the associated gene.
The significant difference in DNA methylation profiles between ovarian cancer cell lines and tumors underscores the need to be cautious in using cell lines as tumor models for molecular studies of ovarian cancer and other cancers. Similarly, the distinct methylation profiles of the different histological types of ovarian tumors reinforces the need to treat the different histologies of ovarian cancer as different diseases, both clinically and in biomarker studies. These data provide a useful resource for future studies, including those of potential tumor progenitor cells, which may help illuminate the etiology and natural history of these cancers.
MicroRNAs (miRNAs) are small regulatory RNAs that are implicated in cancer pathogenesis and have recently shown promise as blood-based biomarkers for cancer detection. Epithelial ovarian cancer is a deadly disease for which improved outcomes could be achieved by successful early detection and enhanced understanding of molecular pathogenesis that leads to improved therapies. A critical step toward these goals is to establish a comprehensive view of miRNAs expressed in epithelial ovarian cancer tissues as well as in normal ovarian surface epithelial cells.
We used massively parallel pyrosequencing (i.e., “454 sequencing”) to discover and characterize novel and known miRNAs expressed in primary cultures of normal human ovarian surface epithelium (HOSE) and in tissue from three of the most common histotypes of ovarian cancer. Deep sequencing of small RNA cDNA libraries derived from normal HOSE and ovarian cancer samples yielded a total of 738,710 high-quality sequence reads, generating comprehensive digital profiles of miRNA expression. Expression profiles for 498 previously annotated miRNAs were delineated and we discovered six novel miRNAs and 39 candidate miRNAs. A set of 124 miRNAs was differentially expressed in normal versus cancer samples and 38 miRNAs were differentially expressed across histologic subtypes of ovarian cancer. Taqman qRT-PCR performed on a subset of miRNAs confirmed results of the sequencing-based study.
This report expands the body of miRNAs known to be expressed in epithelial ovarian cancer and provides a useful resource for future studies of the role of miRNAs in the pathogenesis and early detection of ovarian cancer.
Raman nanoparticle probes are an emerging new class of optical labels for interrogation of physiological and pathological processes in bioassays, cells, and tissues. Although their unique emission signatures are ideal for multiplexing, the full potential of these probes has not been realized because conventional analysis methods are inadequate. We report a novel spectral fitting method that exploits the entire spectral signature to quantitatively extract individual probe signals from multiplex spectra. We evaluate the method in a series of multiplex assays using unconjugated and antibody-conjugated Composite Organic-Inorganic Nanoparticles (COINs). Results show sensitive multiplex detection of small signals (<2% of total signal), and similar detection limits in corresponding 4-plex and singlet plate binding assays. In a triplex assay on formalin-fixed human prostate tissue, two antibody-conjugated COINs and a conventional fluorophore are used to image expression of prostate-specific antigen, cytokeratin-18, and DNA. The spectral analysis method effectively removes tissue autofluorescence and other unknown background, allowing accurate and reproducible imaging (area under ROC curve 0.89+/−0.03) at subcellular spatial resolution. In all assay systems, the error attributable to spectral analysis constitutes ≤2% of total signal. In summary, the spectral fitting method provides (1) quantification of signals from multiplex spectra with overlapping peaks, (2) robust spot-by-spot removal of unknown background, (3) the opportunity to quantitatively assess the analysis error, (4) elimination of operator bias, and (5) simple automation appropriate for high-throughput analysis. The simple implementation and universal applicability of this approach significantly expands the potential of Raman probes for quantitative in-vivo and ex-vivo multiplex analysis.
Composite Organic-Inorganic Nanoparticles (COINs); surface-enhanced Raman scattering (SERS); multiplex assays; tissue imaging; prostate-specific antigen; cytokeratin-18
Animal models greatly facilitate understanding of cancer and importantly, serve pre-clinically for evaluating potential anti-cancer therapies. We developed an invasive orthotopic human glioblastoma multiforme (GBM) mouse model that enables real-time tumor ultrasound imaging and pre-clinical evaluation of anti-neoplastic drugs such as 17-(allylamino)-17-demethoxy geldanamycin (17AAG). Clinically, GBM metastasis rarely happen, but unexpectedly most human GBM tumor cell lines intrinsically possess metastatic potential. We used an experimental lung metastasis assay (ELM) to enrich for metastatic cells and three of four commonly used GBM lines were highly metastatic after repeated ELM selection (M2). These GBM-M2 lines grew more aggressively orthotopically and all showed dramatic multifold increases in IL6, IL8, MCP-1 and GM-CSF expression, cytokines and factors that are associated with GBM and poor prognosis. DBM2 cells, which were derived from the DBTRG-05MG cell line were used to test the efficacy of 17AAG for treatment of intracranial tumors. The DMB2 orthotopic xenografts form highly invasive tumors with areas of central necrosis, vascular hyperplasia and intracranial dissemination. In addition, the orthotopic tumors caused osteolysis and the skull opening correlated to the tumor size, permitting the use of real-time ultrasound imaging to evaluate antitumor drug activity. We show that 17AAG significantly inhibits DBM2 tumor growth with significant drug responses in subcutaneous, lung and orthotopic tumor locations. This model has multiple unique features for investigating the pathobiology of intracranial tumor growth and for monitoring systemic and intracranial responses to antitumor agents.
Elucidation of the repertoire of secreted and cell surface proteins of tumor cells is relevant to molecular diagnostics, tumor imaging and targeted therapies. We have characterized the cell surface proteome and the proteins released into the extra-cellular milieu of three ovarian cancer cell lines, CaOV3, OVCAR3 and ES2 and of ovarian tumor cells enriched from ascites fluid.
Methodology and Findings
To differentiate proteins released into the media from protein constituents of media utilized for culture, cells were grown in the presence of [13C]-labeled lysine. A biotinylation-based approach was used to capture cell surface associated proteins. Our general experimental strategy consisted of fractionation of proteins from individual compartments followed by proteolytic digestion and LC-MS/MS analysis. In total, some 6,400 proteins were identified with high confidence across all specimens and fractions.
Conclusions and Significance
Protein profiles of the cell lines had substantial similarity to the profiles of human ovarian cancer cells from ascites fluid and included protein markers known to be associated with ovarian cancer. Proteomic analysis indicated extensive shedding from extra-cellular domains of proteins expressed on the cell surface, and remarkably high secretion rates for some proteins (nanograms per million cells per hour). Cell surface and secreted proteins identified by in-depth proteomic profiling of ovarian cancer cells may provide new targets for diagnosis and therapy.
In vivo in the prostate gland, basal epithelial cells adhere to laminin 5 (LM5) via α3β1 and α6β4 integrins. When placed in culture primary prostate basal epithelial cells secrete and adhere to their own LM5-rich matrix. Adhesion to LM5 is required for cell survival that is dependent on integrin-mediated, ligand-independent activation of the epidermal growth factor receptor (EGFR) and the cytoplasmic tyrosine kinase Src, but not PI-3K. Integrin-mediated adhesion via α3β1, but not α6β4 integrin, supports cell survival through EGFR by signaling downstream to Erk. PC3 cells, which do not activate EGFR or Erk on LM5-rich matrices, are not dependent on this pathway for survival. PC3 cells are dependent on PI-3K for survival and undergo caspase-dependent death when PI-3K is inhibited. The death induced by inhibition of EGFR or Src in normal primary prostate cells is not mediated through or dependent on caspase activation, but depends on the induction of reactive oxygen species. In addition the presence of an autophagic pathway, maintained by adhesion to matrix through α3β1 and α6β4, prevents the induction of caspases when EGFR or Src is inhibited. Suppression of autophagy is sufficient to induce caspase activation and apoptosis in LM5-adherent primary prostate epithelial cells.
Integrin-mediated adhesion of epithelial cells to extracellular matrix (ECM) proteins induces prolonged tyrosine phosphorylation and partial activation of epidermal growth factor receptor (EGFR) in an integrin-dependent and EGFR ligand-independent manner. Integrin-mediated activation of EGFR in epithelial cells is required for multiple signal transduction events previously shown to be induced by cell adhesion to matrix proteins, including tyrosine phosphorylation of Shc, Cbl, and phospholipase Cγ, and activation of the Ras/Erk and phosphatidylinositol 3′-kinase/Akt signaling pathways. In contrast, activation of focal adhesion kinase, Src, and protein kinase C, adhesion to matrix proteins, cell spreading, migration, and actin cytoskeletal rearrangements are induced independently of EGFR kinase activity. The ability of integrins to induce the activation of EGFR and its subsequent regulation of Erk and Akt activation permitted adhesion-dependent induction of cyclin D1 and p21, Rb phosphorylation, and activation of cdk4 in epithelial cells in the absence of exogenous growth factors. Adhesion of epithelial cells to the ECM failed to efficiently induce degradation of p27, to induce cdk2 activity, or to induce Myc and cyclin A synthesis; subsequently, cells did not progress into S phase. Treatment of ECM-adherent cells with EGF, or overexpression of EGFR or Myc, resulted in restoration of late-G1 cell cycle events and progression into S phase. These results indicate that partial activation of EGFR by integrin receptors plays an important role in mediating events triggered by epithelial cell attachment to ECM; EGFR is necessary for activation of multiple integrin-induced signaling enzymes and sufficient for early events in G1 cell cycle progression. Furthermore, these findings suggest that EGFR or Myc overexpression may provoke ligand-independent proliferation in matrix-attached cells in vivo and could contribute to carcinoma development.
Neutral endopeptidase 24.11 (NEP, CD10) is a cell-surface enzyme expressed by prostatic epithelial cells that cleaves and inactivates neuropeptides implicated in the growth of androgen-independent prostate cancer (PC). NEP substrates such as bombesin and endothelin-1 induce cell migration. We investigated the mechanisms of NEP regulation of cell migration in PC cells, including regulation of phosphorylation on tyrosine of focal adhesion kinase (FAK). Western analyses and cell migration assays revealed an inverse correlation between NEP expression and the levels of FAK phosphorylation and cell migration in PC cell lines. Constitutively expressed NEP, recombinant NEP, and induced NEP expression using a tetracycline-repressive expression system inhibited bombesin- and endothelin-1–stimulated FAK phosphorylation and cell migration. This results from NEP-induced inhibition of neuropeptide-stimulated association of FAK with cSrc protein. Expression of a mutated catalytically inactive NEP protein also resulted in partial inhibition of FAK phosphorylation and cell migration. Coimmunoprecipitation experiments show that NEP associates with tyrosine-phosphorylated Lyn kinase, which then binds the p85 subunit of phosphatidylinositol 3-kinase (PI3-K) resulting in an NEP-Lyn-PI3-K protein complex. This complex competitively blocks FAK-PI3-K interaction, suggesting that NEP protein inhibits cell migration via a protein-protein interaction independent of its catalytic function. These experiments demonstrate that NEP can inhibit FAK phosphorylation on tyrosine and PC cell migration through multiple pathways and suggest that cell migration which contributes to invasion and metastases in PC cells can be regulated by NEP.