Investigation of the effects of a panel of IGFBP-5 mutants on apoptosis in a 3D mammary co-culture system

The mammary gland develops as a rudimentary ductal structure in the mammary fat pad, during pre- and post-pubertal priods. During pregnancy it undergoes massive development resulting in differentiated epithelial cells which ultimately form the secretory alveolar structures that make milk during lactation. At the end of lactation the vast majority of the epithelial cells die by a process of programmed cell death known as apoptosis. The development and survival of the mammary gland depends upon a variety of hormones and growth factors, but it has been shown that the absence of insulin-like growth factor-I (IGF-I) leads to a dramatic impairment of mammary development. IGF-I is now known to be an important survival factor for many different cell types in the body. Surprisingly, the process of cell death at the end of lactation is not accompanied by a decrease in the concentration of IGF-I in the blood. Instead, we have demontrated that the epithelial cells produce a suicide protein, IGF-binding protein-5 (IGFBP-5) which binds to and inhibits the actions of IGF-I. The situation is actually more complex in that we, and others, have shown that IGFBP-5 can act independently of IGF-I, but the manner in which it does so is not yet understood. For example, IGFBP-5 can, independently of IGF-I , activate proteases which are involved in degrading the extracellular environment (a crucial part of the re-modeling of the mammary gland which occurs at the end of lactation). We have already succesfully generated mutated forms of IGFBP-5 in which the IGF-dependent and /independent effects have been separated. These molecules have been studied in preliminary fashion and shown to exhibit novel properties and a principal objective of the current proposal is to use these mutated forms of IGFBP-5 to provide a clearer insight into the mechanisms of action of IGFBP-5 in inducing mammary apoptosis and tissue remodelling. One of the strengths of this proposal lies in the use of a complex 3D model which more closely resembles the in vivo situation than do typical 2D cultures of cells. Our approach involves co-culture of mammary epithelial cells with fat cells (adipocytes) in a collagen/laminin-based extracellular environment. We believe this co-culture to be crucial as, firstly, the mammary epithelium develops as 3D ducts and alveolar structures which are polarised and have an internal cavity (processes which are not achieved with most cell lines). Secondly, the mammary epithelium interacts extensively with the mammary adipocytes, which secrete factors that influence and 'instruct' epithelial cell morphogenesis and differentiation. Our assesment of these co-cultures, and the effects of IGF-I and IGFBP-5 mutants therein, will involve state-of-the-art technologies including confocal microscopy, adenoviral infection with dominant-negative molecules and mutated proteins, and 96-well rtPCR screening approaches. We will also use more classical approaches of immnohistochemistry and western-immunoblotting techniques to examine intra-cellular signalling events. In addition, we will use rapid screening techniques, utilising established cell lines, in order to focus our studies of the more complex 3D cultures. Finally, we will take advantage of transgenic animals expressing the mutant IGFBP-5 molecules specifically in the mammary gland (provided by a separate project) to compare their phenotype with that induced by over-expression of non-mutated IGFBP-5 (this impairs mammary development in vivo). Thus these studies will identify, in vivo and in co-cultures, the relative importance of the IGF-dependent and IGF-independent effects of IGFBP-5 and explore whether this involves changes in cell surface proteins which influence cell survival and migration. We also aim to determine which intracellular signalling pathway(s) are activated by IGFBP-5.

The mammary gland undergoes cycles of development and involution, involving apoptosis. This is induced when the epithelial cells produce IGFBP-5 which inhibits IGF-I-mediated epithelial cell survival. We created transgenic mice expressing IGFBP-5 specifically in the mammary epithelium and showed premature cell death and impaired development. This involved decreased IGF-signalling, PI3-kinase activity, decreased expression of pro-survival members of the Bcl-2 family and increased caspase-3 activity. IGFBP-5 also possesses IGF-independent effects which influence activation of proteases and cell migration. The effects of IGFBP-5 mimic those of TGFB-1 which has been shown to up-regulate IGFBP-5 expression. Furthermore, their temporal and spatial expression patterns in the mammary gland are very similar. We will test the hypothesis that IGFBP-5 mediates the effects of TGFB-1 using a shRNA-IGFBP-5 construct. The project will also seek to dissociate individual properties of IGFBP-5. The IGF-independent effects of IGFBP-5 reside principally in a heparin-binding domain in the C-terminus, whereas its IGF-binding properties lie principally in the N-terminus. We have produced IGFBP-5 mutants in which IGF- and heparin-binding have been selectively deleted, in order to compare their biological properties with wt-IGFBP-5. The strength of this proposal lies in a 3D model of the mammary gland which re-capitulates many of the features of the normal gland. We will also utilise transgenic mice expressing these two IGFBP-5 mutants, to compare their phenotype with our existing wt-IGFBP-5 transgenic mouse. Thus our principal objectives will be to produce, express and assess by Biosensor analysis, interactions of the mutants with IGFs, heparin and various ECM molecules and to determine their effects in 3-D 'mammosphere' cultures, in 3 human mammary epithelial cell lines and in transgenic mice.