David R. Rowley

REACTIVE STROMA IN PROSTATE CANCER PROGRESSION

PURPOSE The development of an altered stromal microenvironment in response to carcinoma is a common feature of many tumors. We reviewed the literature describing characteristics of reactive stroma, how reactive stroma affects cancer progression and how carcinoma regulates reactive stroma. Moreover, we present a hypothesis of reactive stroma in prostate cancer and discuss how the biology of reactive stroma may be used in novel diagnostic and therapeutic approaches. MATERIALS AND METHODS An extensive literature search was performed to review reports of the general features of wound repair stroma, general stromal responses to carcinoma, and stromal biology of normal and prostate cancer tissues. These studies were analyzed and a reactive stroma hypothesis in prostate cancer was developed. RESULTS Modifications to the stroma of breast, colon and prostate tumors parallel the generation of granulation tissue in wound repair. These changes include stromal cell phenotypic switching, extracellular matrix remodeling and angiogenesis induction. Therefore, it is predicted that a modified wound healing response induces the formation of reactive stroma in cancer to create a tumor promoting environment. Based on its role in wound repair and its over expression in prostate cancer, transforming growth factor-beta stands out as a potential regulator of reactive stroma. CONCLUSIONS Reactive stroma in prostate cancer and granulation tissue in wound repair show similar biological responses and processes that are predicted to promote cancer progression. Further identification of specific functional and regulatory mechanisms in prostate cancer reactive stroma may aid in the use of reactive stroma for novel diagnostic and therapeutic approaches.

Stromal Expression of Connective Tissue Growth Factor Promotes Angiogenesis and Prostate Cancer Tumorigenesis

Our previous studies have defined reactive stroma in human prostate cancer and have developed the differential reactive stroma (DRS) xenograft model to evaluate mechanisms of how reactive stroma promotes carcinoma tumorigenesis. Analysis of several normal human prostate stromal cell lines in the DRS model showed that some rapidly promoted LNCaP prostate carcinoma cell tumorigenesis and others had no effect. These differential effects were due, in part, to elevated angiogenesis and were transforming growth factor (TGF)-beta1 mediated. The present study was conducted to identify and evaluate candidate genes expressed in prostate stromal cells responsible for this differential tumor-promoting activity. Differential cDNA microarray analyses showed that connective tissue growth factor (CTGF) was expressed at low levels in nontumor-promoting prostate stromal cells and was constitutively expressed in tumor-promoting prostate stromal cells. TGF-beta1 stimulated CTGF message expression in nontumor-promoting prostate stromal cells. To evaluate the role of stromal-expressed CTGF in tumor progression, either engineered mouse prostate stromal fibroblasts expressing retroviral-introduced CTGF or 3T3 fibroblasts engineered with mifepristone-regulated CTGF were combined with LNCaP human prostate cancer cells in the DRS xenograft tumor model under different extracellular matrix conditions. Expression of CTGF in tumor-reactive stroma induced significant increases in microvessel density and xenograft tumor growth under several conditions tested. These data suggest that CTGF is a downstream mediator of TGF-beta1 action in cancer-associated reactive stroma and is likely to be one of the key regulators of angiogenesis in the tumor-reactive stromal microenvironment.

What Might A Stromal Response Mean to Prostate Cancer Progression

Many similarities exist between the stroma at sites of wound repair and reactive stroma in cancer. Common features include an elevated stromal cell proliferation, altered expression of matrix components, elevated expression of TGFβ-1, neovascularization, and expression of several common stromal markers. In addition, proliferative stromal cells at these sites generally express myodifferentiation markers. A comparison between the many common features and the biologically active molecules observed in reactive stroma in carcinoma and reactive stroma in wound repair is discussed in this review. An extended analysis of the literature suggests a functional link between mechanisms in wound repair response and the stromal reaction in many cancers including prostate cancer. We propose in this review, that the fundamental mechanisms of stroma in providing a rapid response to altered homeostasis in wounding, also provides for a tumor-regulating stromal microenvironment in cancer. The functional consequences of this stromal response to carcinoma progression and how the stromal response might be used in extended diagnosis and in therapeutic approaches are discussed.

Growth and Survival Mechanisms Associated with Perineural Invasion in Prostate Cancer

Perineural invasion (PNI) is the major mechanism of prostate cancer spread outside the prostate. Apoptotic and proliferation indices were determined in PNI cells using the PNI in vitro model and human PNI in tissue microarrays. RNA was extracted from the PNI model and controls and evaluated by cDNA microarray analysis. Differential expression of candidate genes was confirmed by real-time quantitative PCR, fluorescence, and immunohistochemistry using tissue microarrays. Genistein and BAY 11-7085 were added to the supernatant of cocultures and controls in microchamber cultures. The significance of nuclear factor κB (NFκB) nuclear translocation in human PNI was analyzed using Kaplan-Meier analysis. An increase in proliferation and a decrease in apoptosis were observed in human PNI cells and the PNI model as compared with controls. Three of 15 genes up-regulated in the cDNA microarray were involved in the apoptosis signaling pathway (NFκB), and its downstream targets defender against cell death 1 and PIM-2. The increase was corroborated by real-time quantitative PCR and immunofluorescence. NFκB nuclear translocation was seen in the in vitro model and human tissues, where strong nuclear expression was associated with a decrease in recurrence-free survival. Addition of genistein and BAY 11-7085 resulted in a decrease in NFκB, PIM-2 and defender against cell death 1 as well as a reversal of the inhibition of apoptosis. This is the first description of a biological mechanism and functional significance of PNI. Cancer cells in a perineural location acquire a survival and growth advantage using a NFκB survival pathway. Targeting PNI might help detain local spread of the tumor and influence survival.

Production and characterization of an antibody to myosin light chain kinase and intracellular localization of the enzyme

A specific precipitating antibody against chicken gizzard myosin light chain kinase (MLCK) has been produced in rabbits. The antibody inhibited enzyme activity in a dose-dependent manner with 11 moles of antibody required to inhibit 80% of the activity of one mole MLCK. Crude homogenates from various chicken tissues were analyzed by SDS-polyacrylamide gel electrophoresis, and the proteins were transferred onto nitrocellulose sheets and reacted with antibody. In each case, the antibody bound to only one protein with a molecular weight of approximately 130,000. These data suggest the MLCK present in all types of muscle as well as non-muscle tissues is of the same molecular weight. Indirect immunofluorescent microscopy of non-muscle tissue culture cells revealed MLCK to be localized in the spindle apparatus and midbody of mitotic cells and on the stress fibers and in the nucleolus of interphase cells. The nucleolar localization was confirmed by electron microscopy and shown to be restricted to the fibrillar region. In myofibrils isolated from skeletal and cardiac muscle, anti-MLCK decorated the actin-containing I bands of the sarcomere. These results are consistent with the suggestion that MLCK and calmodulin are intermediates in the regulation of cell motility by calcium.

Global gene expression analysis of reactive stroma in prostate cancer

Purpose: Marked reactive stroma formation, designated as grade 3 reactive stroma, is associated with poor outcome in clinically localized prostate cancer. To understand the biological processes and signaling mechanisms underlying the formation of such reactive stroma, we carried out microarray gene expression analysis of laser-captured reactive stroma and matched normal stroma. Experimental Design: Seventeen cases of reactive stroma grade 3 cancer were used to laser-capture tumor and normal stroma. Expression analysis was carried out using Agilent 44K arrays. Up-regulation of selected genes was confirmed by quantitative reverse transcription-PCR. Expression data was analyzed to identify significantly up- and down-regulated genes, and gene ontology analysis was used to define pathways altered in reactive stroma. Results: A total of 544 unique genes were significantly higher in the reactive stroma and 606 unique genes were lower. Gene ontology analysis revealed significant alterations in a number of novel processes in prostate cancer reactive stroma, including neurogenesis, axonogenesis, and the DNA damage/repair pathways, as well as evidence of increases in stem cells in prostate cancer reactive stroma. Conclusions: Formation of reactive stroma in prostate cancer is a dynamic process characterized by significant alterations in growth factor and signal transduction pathways and formation of new structures, including nerves and axons.

Cancer-Related Axonogenesis and Neurogenesis in Prostate Cancer

Purpose: Perineural invasion is the only interaction between cancer cells and nerves studied to date. It is a symbiotic relationship between cancer and nerves that results in growth advantage for both. In this article, we present data on a novel biological phenomenon, cancer-related axonogenesis and neurogenesis. Experimental Design: We identify spatial and temporal associations between increased nerve density and preneoplastic and neoplastic lesions of the human prostate. Results: Nerve density was increased in cancer areas as well as in preneoplastic lesions compared with controls. Two- and three-dimensional reconstructions of entire prostates confirmed axonogenesis in human tumors. Furthermore, patients with prostate cancer had increased numbers of neurons in their prostatic ganglia compared with controls, corroborating neurogenesis. Finally, two in vitro models confirmed that cancer cells, particularly when interacting with nerves in perineural invasion, induce neurite outgrowth in prostate cancer. Neurogenesis is correlated with features of aggressive prostate cancer and with recurrence in prostate cancer. We also present a putative regulatory mechanism based on semaphorin 4F (S4F). S4F is overexpressed in cancers cells in the perineural in vitro model. Overexpression of S4F in prostate cancer cells induces neurogenesis in the N1E-115 neurogenesis assay and S4F inhibition by small interfering RNA blocks this effect. Conclusions: This is the first description of cancer-related neurogenesis and its putative regulatory mechanism.

Antitumor effects of chimeric receptor engineered human T cells directed to tumor stroma.

Cancer-associated fibroblasts (CAFs), the principle component of the tumor-associated stroma, form a highly protumorigenic and immunosuppressive microenvironment that mediates therapeutic resistance. Co-targeting CAFs in addition to cancer cells may therefore augment the antitumor response. Fibroblast activation protein-α (FAP), a type 2 dipeptidyl peptidase, is expressed on CAFs in a majority of solid tumors making it an attractive immunotherapeutic target. To target FAP-positive CAFs in the tumor-associated stroma, we genetically modified T cells to express a FAP-specific chimeric antigen receptor (CAR). The resulting FAP-specific T cells recognized and killed FAP-positive target cells as determined by proinflammatory cytokine release and target cell lysis. In an established A549 lung cancer model, adoptive transfer of FAP-specific T cells significantly reduced FAP-positive stromal cells, with a concomitant decrease in tumor growth. Combining these FAP-specific T cells with T cells that targeted the EphA2 antigen on the A549 cancer cells themselves significantly enhanced overall antitumor activity and conferred a survival advantage compared to either alone. Our study underscores the value of co-targeting both CAFs and cancer cells to increase the benefits of T-cell immunotherapy for solid tumors.

Fibroblast growth factor-2 mediates transforming growth factor- β action in prostate cancer reactive stroma

Transforming growth factor-β (TGF-β) is overexpressed at sites of wound repair and in most adenocarcinomas including prostate cancer. In stromal tissues, TGF-β regulates cell proliferation, phenotype and matrix synthesis. To address mechanisms of TGF-β action in cancer-associated reactive stroma, we developed prostate stromal cells null for TGF-β receptor II (TβRII) or engineered to express a dominant-negative Smad3 to attenuate TGF-β signaling. The differential reactive stroma (DRS) xenograft model was used to evaluate altered stromal TGF-β signaling on LNCaP tumor progression. LNCaP xenograft tumors constructed with TβRII null or dominant-negative Smad3 stromal cells exhibited a significant reduction in mass and microvessel density relative to controls. Additionally, decreased cellular fibroblast growth factor-2 (FGF-2) immunostaining was associated with attenuated TGF-β signaling in stroma. In vitro, TGF-β stimulated stromal FGF-2 expression and release. However, stromal cells with attenuated TGF-β signaling were refractory to TGF-β-stimulated FGF-2 expression and release. Re-expression of FGF-2 in these stromal cells in DRS xenografts resulted in restored tumor mass and microvessel density. In summary, these data show that TGF-β signaling in reactive stroma is angiogenic and tumor promoting and that this effect is mediated in part through a TβRII/Smad3-dependent upregulation of FGF-2 expression and release.

Elevated epithelial expression of interleukin-8 correlates with myofibroblast reactive stroma in benign prostatic hyperplasia.

OBJECTIVES Numerous inflammatory diseases display elevated interleukin (IL)-8, and most are associated with a reactive stroma. IL-8 expression is also elevated in benign prostatic hyperplasia (BPH), yet little is known about reactive stroma in BPH. Whether a reactive stroma response exists in BPH, whether this correlates with elevated IL-8, and whether IL-8 can induce a reactive stroma phenotype have not been determined. This study was designed to specifically address these issues. METHODS Normal prostate transition zone tissue and BPH specimens, as identified by the Baylor College of Medicine pathology department, were examined by quantitative immunohistochemistry to correlate IL-8, smooth muscle alpha-actin, vimentin, calponin, and tenascin-C. In addition, human prostate stromal cell cultures were used to evaluate the effect of IL-8 on the expression of reactive stroma biomarkers. RESULTS BPH nodules exhibited elevated epithelial IL-8 immunoreactivity, and this correlated with elevated smooth muscle alpha-actin, reduced calponin, and altered deposition of tenascin-C, relative to the normal prostate transition zone tissue (P <0.05). Multiple vimentin-positive prostate stromal fibroblast cultures were induced by IL-8 to also co-express smooth muscle alpha-actin and tenascin-C, typical of a reactive stroma myofibroblast phenotype. CONCLUSIONS These data show that BPH reactive stroma is fundamentally different from normal prostate fibromuscular stroma and is typified by the emergence of a reactive stroma myofibroblast phenotype. This reactive stroma pattern correlated spatially with IL-8 elevation in adjacent epithelium. Additionally, IL-8 induced expression of myofibroblast markers in human prostate fibroblasts in vitro. These results suggest that IL-8 acts as a regulator of BPH reactive stroma and is therefore a potential therapeutic target.