Prostate cancer specific integrin alphavbeta3 modulates bone metastatic growth and tissue remodeling

Abstract

The management of pain and morbidity due to the spreading and growth of cancer within bone remains to be a paramount problem in clinical care. Cancer cells actively transform bone, however, the molecular requirements and mechanisms of this process remain unclear. This study shows that functional modulation of the alphavbeta3 integrin receptor in prostate cancer cells is required for progression within bone and determines tumor-induced bone tissue transformation. Using histology and quantitative microCT analysis, we show that alphavbeta3 integrin is required not only for tumor growth within the bone but for tumor-induced bone gain, a response resembling bone lesions in prostate cancer patients. Expression of normal, fully functional alphavbeta3 enabled tumor growth in bone (incidence: 4/4), whereas alphavbeta3 (-), inactive or constitutively active mutants of alphavbeta3 did not (incidence: 0/4, 0/6 and 1/7, respectively) within a 35-day-period. This response appeared to be bone-specific in comparison to the subcutis where tumor incidence was greater than 60% for all groups. Interestingly, bone residing prostate cancer cells expressing normal or dis-regulated alphavbeta3 (either inactive of constitutively active), but not those lacking beta3 promoted bone gain or afforded protection from bone loss in the presence or absence of histologically detectable tumor 35 days following implantation. As bone is replete with ligands for beta3 integrin, we next demonstrated that alphavbeta3 integrin activation on tumor cells is essential for the recognition of key bone-specific matrix proteins. As a result, prostate cancer cells expressing fully functional but not dis-regulated alphavbeta3 integrin are able to control their own adherence and migration to bone matrix, functions that facilitate tumor growth and control bone lesion development.

Figure 1

In vivo bone imaging using microCT. Eight-week-old-male NOD CB17PRK Scid/J mice (Jackson Laboratories, Bar Harbor, ME, USA) were injected intratibially with C4–2 cells (2.5 × 10⁵ in 20 μl PBS). Contralateral tibiae were injected with PBS alone. Mice were anesthetized and secured on the rotating platform of a custom-designed microCT system (Latson et al., 2003). Three-dimensional reconstructions of 2D image stacks of injected tibiae 1 day and 35 days post-implantation of C4–2 cells are shown. The foreground is clipped to give a sagittal view of the bone cavity from the medial side. Insets depict transverse slices (medial is down, anterior is left) corresponding to the plane designated by white arrows.

Figure 2

MicroCT analysis of changes in bone structural indices. Tibial segmentation from surrounding bone, volume enhancement and delineation of volumes-of-interest (VOIs) were performed using open-source 3D visualization software (VolSuite, Ohio Supercomputer Center, Columbus, OH, USA). For each sample, a plane perpendicular to the z axis/tibial shaft was generated and placed at the base of the growth plate. A second, parallel plane was defined ∼1.0 mm below the first and the entire volume was clipped to this VOI for quantitative analysis. (a) Changes in bone fraction of tibiae over a 35-day-period following injection with C4–2 prostate cancer cells lacking or expressing αvβ3 integrin in different functional states. *P<0.05 vs PBS control injections within the same group. (b) Changes in the number of trabeculae in the same bones as in (a). **P<0.01 vs PBS control injections within the same group or change in trabecular number of control vs C4–2 injected within the C4–2 αvβ3 WT group compared to the same change within the αvβ3 (—) group.

Figure 3

Impairment of tumor growth as a consequence of αvβ3 function is specific to the bone microenvironment. Intratibial injections were as described in Figure 1. Alternatively, mice were injected subcutaneously (sc) with matrigel suspensions of 1×10⁶ C4–2 cells. (a) Left column: representative H&E histology of injected tibiae. Note that the majority of normal marrow (M) has been displaced by tumor (T) in the C4–2 αvβ3 WT injected tibia (second row). Tumor incidence is shown in insets. Original magnification of ×20. Right column: tumor presence was confirmed immunohistochemically for the presence of PSMA. Asterisks indicate cortical bone. Original magnification of ×40. (b) Left column. Gross tumor analysis 4 weeks following injection of C4–2 cells. Tumor incidence is indicated in the inset. Right column: hematoylxin and eosin (H&E) histology of sc tumor sections from each injection group. Original magnification of ×20. Histopathology of sc tumors was performed as described previously (De et al., 2003). Injected tibiae were fixed and decalcified overnight in Decalcifier (Surgipath, Richmond, IL, USA), cut longitudinally and embedded in paraffin. Sections (5 μm) were then stained with H&E or subjected to immunohistochemistry with anti-human PSMA (4D8; Northwest Biotherapeutics, Bothell, WA, USA) or a polyclonal rabbit anti-bovine cytokeratin antibody (Z0622; Dako, Carpinteria, CA, USA) followed by incubation with biotinylated secondaries and developed.