An investigation into the function of Cysteine-Rich Secretory Protein 3

Cysteine-Rich Secretory Protein 3 (CRISP3) is found predominantly in the salivary gland and granulocytes of healthy subjects, but is significantly up-regulated in the malignant prostate, where it represents the most over-expressed transcript. While the expression pattern of CRISP3, its up-regulation in several inflammatory pathologies and the immunosuppressive role of related proteins in other species suggest a role for human CRISP3 in the regulation of the immune response, no definitive function has been demonstrated. An anti-proliferative role of CRISP3 in prostate cancer has also been proposed based on in vitro transfection experiments. Progress in understanding CRISP3 function has been slow due partly to the fact that CRISP3 is normally expressed at low levels and has a complex tertiary structure defined by eight conserved disulfide bonds, thus limiting attempts at purification from both native and recombinant sources for use in functional assays. Efforts to purify CRISP3 from bacterial expression systems have invariably given rise to misfolded protein aggregates. The aims of the work presented in this thesis were threefold; to establish an efficient system of purification of CRISP3 in its native conformation, to investigate the effects of CRISP3 on prostatic epithelial cell proliferation, and to define the effect of Crisp3 knockout in the Hi-Myc model of murine prostate cancer. In Chapter 2, I present a two-step system for recombinant CRISP3 expression and purification involving cation-exchange and size exclusion chromatography based on a mammalian expression system. This protocol provided milligram quantities of recombinant CRISP3. The presence of appropriate post-translational modifications and the interaction of the purified protein with a known CRISP3 binding partner, which was abolished with CRISP3 denaturation, indicated that the recombinant protein was likely to be in its native conformation. A comparative study of N-glycosylation sites and types within human and mouse CRISP3 was undertaken, and revealed striking interspecies differences in post-translational modification. This represents the first study of mouse CRISP3 and characterization of its glycosylation pattern. Our recombinant CRISP3 preparation was used alongside transfection experiments to show, in Chapter 3, that contrary to a published report, neither extracellular treatment with, nor intracellular overexpression of, CRISP3 directly affected the proliferation of PC3 prostate cancer cells. Similarly, shRNA knockdown of Crisp3 had no effect on prostate cell proliferation. In order to investigate the in vivo function of CRISP3 in prostate malignancy, we made use of the murine Hi-Myc model of prostate cancer, which relies on the induced prostate-specific overexpression of the human c-Myc oncogene to initiate cancer, which typically becomes invasive between 6-12 months of age. A two-generational breeding programme was used to generate Crisp3-/- and Crisp3+/+ Hi-Myc mice. Cancer progression in the two genotypes was compared with histology and markers of proliferation, tumour burden, invasion, and tumour leukocyte infiltration. CRISP3 expression was absent in the normal wild type prostate, but underwent significant up-regulation between 5 to 7 months in the Hi-Myc lateral prostate. This up-regulation coincided with accelerated progression to an invasive cancer phenotype by 7 months in the Crisp3+/+ animals. In contrast, Crisp3 knockout totally suppressed progression from duct-contained (carcinoma in situ) to invasive prostate cancer in Crisp3-/- mice at 7 months. Hence, the experiments presented in this thesis demonstrate for the first time that CRISP3 is pro-tumourigenic in the malignant prostate. Based on immunhistochemistry with the proliferation marker Ki67 and the work contained in Chapter 3, CRISP3 is unlikely to directly affect the proliferation of prostatic epithelium. Rather, preliminary data suggest that inflammatory mediators such as TGF-β can induce CRISP3 up-regulation, which in turn may act on both the prostatic epithelium to promote an epithelial to mesenchymal transition, and on tumour-infiltrating immune cells to condition a tolerogenic immune phenotype. This work therefore represents an important advance in the field’s understanding of the role of CRISP3 in prostate cancer