It is estimated that in the year 2000, some 6 million people worldwide died as a result of cancer. Based on current rates, it is predicted that in 2050 this number will rise to 16 million, and 23 million new cases will be identified.
Despite intensive research, medicine still lacks the means to effectively treat many tumour forming cancers, and this is reflected in the number of cancer related deaths. A major impediment to progress is a lack of appropriate in vitro cell culture models of solid tumours. Commonly, simple monolayer cultures of isolated tumour derived cells are employed. These cultures fail to account for some of the most basic features of tumours, and as a consequence are of limited clinical relevance.
In the 1970's, the multicellular tumour spheroid was introduced as an in vitro model of intermediate complexity between monolayers and in vivo tumours. These spherical cellular aggregates exhibit a physiochemical environment similar to that observed in tumours, a corresponding heterogeneous population of cells, and a response to anti-tumour therapies that correlates well with clinical findings. Since their first introduction, numerous studies have confirmed the tumour-like nature of multicellular spheroids (MCS) and their utility in cancer research has been well demonstrated. However, many investigators today still employ monolayer cultures in studies for which they are inappropriate and MCS would be far better suited. There are several possible reasons as to why this is so, among them the fact that MCS cultivation is more difficult, and data analysis and interpretation is more complex than for the more familiar monolayer.
Current methods of MCS cultivation result in a population of widely distributed size, may fail for a given cell fine, provide little control over composition, and often include the use of artificial matrix materials that can alter cellular behaviour. New cultivation methods that address these issues are required.
This thesis describes the application of the hanging drop method for the cultivation of MCS. Using this method it has been possible to cultivate large numbers of homogeneously sized MCS from 40 different cell lines. These cultures required little effort, yielding well shaped MCS within days of initiation. The simplicity of this method makes effective controlled cultivation of MCS accessible to any reasonably equipped cell culture facility, without the need for specialist apparatus or personnel.
While it is known that tumours are less sensitive to anti-cancer agents than monolayers, the underlying mechanisms of this phenomenon are poorly understood. Initially it was believed that resistance arose due to poor penetration of drugs, reduced oxygen concentrations, and the presence of quiescent cell populations. More recent evidence indicates that fundamental changes in cellular phenotype arising as a result of 3D growth are responsible for this resistance, and this effect has been termed multicellular resistance (MCR). Underlying these changes in cellular phenotype, are changes in the expression of multiple genes and gene products.
Using colon carcinoma derived MCS generated by the hanging drop method, 42 genes that are differentially expressed in MCS compared to monolayers were identified. Of these genes, 3 might have particular relevance to MCR. The functions of these genes (S100A4, SKIP3 and p48) suggest three different mechanisms of resistance: modification of the intracellular environment, suppression of apoptosis, and enhanced DNA damage repair.
Not only is an increased understanding of the clinical features of tumours required, but also of the processes by which they develop. Breast cancer is an excellent example of a tissue that, due its nature, is prone to malignant transformation. In vitro models with which to study this transformation have been of great value in understanding the events that lead to the development of breast cancer. Unfortunately, a reliance on matrix substitutes such as matrigel makes these cultures expensive, and might also subtly influence the behaviour of cells in ways that are not yet recognised.
In hanging drop cultures, freshly isolated murine mammary epithelial cells collected into small aggregates resembling those observed in matrigel cultures. Optical sectioning of these structures revealed a single layer of cells surrounding a central lumen, with a layer of laminin deposited at the outer surface. When prolactin was included in the culture medium, milk proteins could be detected within the lumen. These observations demonstrate a polarised cell morphology and functional phenotype. Thus, it is possible to effectively and cheaply cultivate functional primary mammary organoids in the absence of matrix substitutes, using the hanging drop method.
This ability of cells to self-organise in hanging drop cultures also facilitates the development of mixed cell population tissue models, being more representative of the complex in vivo situation. Cells other than the malignant population contribute to the overall tumour phenotype, and due to the frequency and magnitude of therapeutic resistance to current treatments, new strategies that target these 'other cells' are being pursued. One promising approach is targeting the process of tumour angiogenesis. However, this pursuit is hampered by a lack of in vitro angiogenesis models that are representative of the in vivo process.
Existing in vitro approaches typically rely on the cultivation of endothelial cells on, or within, some form of matrix substitute. While they have been useful for studies of individual aspects of angiogenesis, these models lack a tumour component and fail to recapture the interactions that occur between tumours and endothelial cells.
Using the hanging drop method, it was possible to generate microvascularised MCS by co-cultivation with isolated human umbilical vein endothelial cells (HUVEC). Upon introduction to hanging drops, HUVEC migrate into the MCS where they establish tubular networks. The precise behaviour of HUVEC in this culture format varied for MCS composed of different cell lines, possibly reflecting tissue specific differences. These constructs are straightforward to cultivate, and have potential for use in the screening and characterisation of anti-angiogenic agents.
This thesis demonstrates that using the hanging drop method, MCS can be readily cultivated for use in traditional and novel applications. This method is an enabling technology, increasing accessibility to, and extending the application of MCS as an in vitro tissue model.