eHsp90 as a modulator of the tumor microenvironment
Highlights
-
Molecular markers are unable to accurately predict risk for prostate cancer progression.
-
We are exploring whether molecular indicators in the tumor stroma may signify progression risk
-
eHsp90 initiates molecular events consistent with a CAF-like phenotype.
-
Tumor eHsp90 may therefore signal in a paracrine manner to deregulate tumor-stromal communication and support tumor progression.
-
eHsp90 may deregulate stromal microRNA expression to drive CAF-like behavior
Background and Significance
Prostate cancer is generally a slow-growing tumor type. Early detection of localized, indolent tumors portends a good prognosis. Although most early tumors will not progress, it remains a significant challenge to identify the patients with tumors that will become invasive and disseminate. There is currently no cure for metastatic prostate cancer, which ranks as the second highest cause of cancer lethality for males. While EMT events are thought to play a role in this switch towards invasive and metastatic behavior, it is also thought that changes within the tumor microenvironment contribute to this progression. Currently, there are no clinical biomarkers that are able to accurately identify at risk patients. Most genomic-based studies have focused on genetic changes occurring in the cancer epithelial cells. However, the complex genetic heterogeneity of this disease has confounded attempts to define a genetic signature for progression. We believe that the tumor environment may hold clues for progression risk.
Salient Findings
Although soluble tumor factors promote the conversion of normal fibroblasts into CAFs, the mechanism for this reprogramming is not well understood. We demonstrated that prostate tumor eHsp90 signals in an autocrine pathway to maintain the aggressive nature of tumor cells. Soluble tumor factors may have effects upon multiple cell types. Therefore, we set out to test whether eHsp90 may also function in a paracrine pathway to modulate the properties of stromal cell types. To investigate this, we utilized prostate stromal fibroblasts (PrSFs). Hsp90 protein was added to PrSFs (to mimic eHsp90 secretion) following a scratch wound assay to evaluate cell motility. As shown, eHsp90 dramatically increased PrSF cell motility (Fig. 10A). We then investigated whether eHsp90 induced an inflammatory phenotype common to CAF-like cells. Hsp90 protein was added to PrSF cells and the media was subsequently analyzed for expression of known tumor-promoting inflammatory cytokines IL-6 and IL-8. As shown, eHsp90 induced the expression of these cytokines (Fig. 10B).
Ongoing and Future Research
We are working to understand how eHsp90 initiates the promotion of CAF-like behavior. We are also evaluating the clinical relevance of this finding, and determining its prevalence. Towards this goal, we are investigating whether an eHsp90-initiated CAF-like phenotype may be found in patient samples. We are working with a collaborator (Dr. Hayward, Vanderbilt University) to obtain matched normal prostatic fibroblasts (NPFs) and CAFs (Fig. 12A).
Moreover, eHsp90-regulated miRs (green circle) demonstrate significant overlap with the patient derived trends. This indicates that eHsp90 regulates miRs that are associated with CAF-like behavior in patient samples. Identification and functional validation of these miRs in animal models may provide biomarkers capable of discerning the relative risk of progression.
Project Two
Thus, eHsp90 imparts properties consistent with CAF-like behavior. A model for eHsp90 as a modulator of tumor-stromal communication is shown (Fig. 11). This findings have been recently submitted (Bohonowych JE et al. Prostate, 2013).
We are particularly interested in Cancer-Associated Fibroblasts (CAFs). CAFs are stromal fibroblasts that acquire the ability to support tumor growth and metastasis. CAFs can be induced following incubation of normal fibroblasts with the soluble milieu from cancer cells. CAFs acquire the ability to secrete chemokines and cytokines that support tumor growth and can activate tumor cell EMT. The presence of an inflammatory environment, and CAF-like cells, is observed in most solid tumors. If CAFs can be targeted, it may be possible to attenuate the tumor-supportive environment that contributes to progression.
Epithelial tumor cells do not act alone. It is becoming clear that they influence the tumor microenvironment, consisting primarily of host stromal cell types. This tumor-stromal communication is thought to play a major role in cancer progression (Fig. 9).
The isolation of CAF-like cells is accomplished by isolating fibroblasts in close proximity to the tumor cells, whereas NPFs are isolated from adjacent benign tumor tissue. The CAF-like activity is validated by its ability to support epithelial transformation in animal models. We are evaluating whether CAFs exhibit differentially expressed microRNAs (miRs). We are evaluating miRs because they are known to control large numbers of genes and to regulate major programs. We are utilizing our genetic model of directed eHsp90 secretion (ARCaPE-eHsp90) to determine whether eHsp90 regulates miRs in common with CAF-specific miRs. As shown via Venn diagram, a subset of CAF-specific miRs are shared between two evaluated patients (blue and yellow circles) (Fig. 12B).
