**1. Introduction**

136 Bioluminescence – Recent Advances in Oceanic Measurements and Laboratory Applications

White, P., Squirrell, D., Arnaud, P., Lowe, C. and Murray, J. (1996). Improved

White, E.; Worther, H., Seliger, H. & McElroy, W. (1966). Amino Analogs of Firefly Luciferin

Willey, T.L., Squirrell, D. and White, P. (2001). Design and Selection of Firefly Luciferases

Wood K. (1998). The Chemistry of Bioluminescent Reporter Assays. *Promega Notes,* No.65,

Woodroofe, C.; Shultz, J., Wood, M., Cali, J., Daily, W., Meisenheimer, P. & Laubert, D.

Zinn K.; Chaudhuri, T., Szafran, A., O'Quinn, D., Weaver, C., Dugger, K., Lamar, D.,

at Position 354. *Biochemical Journal,* No.319, pp. 343-350

and Activity Thereof. *JACS,* No. 88, pp. 2015-2018

World Scientific, Singapore. ISBN: 981024679X

in Small Animals. *ILAR J,* No.49, pp. 103-115

*Biochemistry,* No. 47, 10383-10393

pp. 14-21

Thermostability of the North American Firefly Luciferase: Saturation Mutagenesis

with Novel in vivo and in vitro Properties. In *Proc. 11th Int. Symp. Biolum. Chemilum*. Case, J.; Herring, P., Robinson, B., Haddock, S., Kricka, L. & Stanley, P. pp. 201-204,

(2008). N-Alkylated 6'-Aminoluciferins are Bioluminescent Substrates for Ultra-Glo and Quantilum Luciferase: New Potential Scaffolds for Bioluminescent Assays.

Kesterson, R., Wang, X. & Frank, S. (2008). Noninvasive Bioluminescence Imaging

In vitro studies have offered vast insight into much of cancer biology, however, it is widely accepted that cell based assays are unable to provide a complete picture when attempting to understand the dynamic nature of cancer as it behaves in situ. Critical processes to cancer progression such as angiogenesis, metastasis and response to treatment, rely on complex interactions between tumor cells and their microenvironment. To overcome this challenge, xenografts have been widely used to study cancer biology within the context of a whole organism since as early as the 1950's. These models rely on use of murine (syngeneic) and human (allogeneic) tumor cell lines injected subcutaneously into a rodent host. The subcutaneous animal model has been a valuable tool in the study of cancer and has directly led to the validation of many of the anticancer agents which benefit patients today. Subcutaneous tumor models are easy to implement and monitor due to the accessibility of the tumor tissue. Evaluation of subcutaneous tumors involves calliper measurements of tumor size (width, length and/or height), which are then used to define tumor volume. However, like cell-based assays, subcutaneous tumor models have proven to be poor predictors of therapeutic activity in patients and this is likely due to the reliance on cell lines which when inoculated subcutaneously develop tumors that poorly mimic the biological behaviour of human disease. Cancers arise slowly and evolve into a heterogeneous structure both in terms of cellular composition (host cells and tumor cells) and microenvironment (vascularization and transient regions of hypoxia and nutrient stress). Subcutaneous xenografts are implanted in microenvironments that will be remarkably different from the tissue of origin. This means subcutaneous tumor cells do not receive the same signals from the stroma that influence immunity, angiogenesis and metastasis; all factors that impact tumor progression and response to therapeutic interventions. Although the initiators and drivers of tumors in humans remain poorly understood, it is generally accepted that following initiation, endogenous disease progresses into a primary tumor which in time can invade surrounding tissues. The latter process involves both extravasation and intravasation

Bioluminescence Applications in Preclinical Oncology Research 139

increasingly popular in part because of the accessibility of imaging tools and because the method is extremely sensitive with a capability of detecting as few as 10 tumor cells in a live animal. BLI provides a non-invasive, semi-quantitative approach to localizing small tumors, following growth and metastasis and monitoring tumor response to treatment in the same animal longitudinally. This non-invasive determination of tumor burden over time reduces the numbers of animals required for experiments and provides information on the various

The first experiments using BLI for monitoring tumor phenotypes and response to therapy were performed by assessing luciferase activity as a measure of metabolism as described in section 6.0. Recently, more complex studies have been designed. For example, our group used BLI to track the development of an experimental metastatic model of breast cancer after an intracardiac injection of tumor cells. Further, we evaluated the use of an existing and clinically relevant drug to treat orthotopic, metastatic and ascities disease and correlated changes in tumor burden as measured by BLI to pharmacokinetic data in the same animals. The utility of BLI in assessing drug efficacy is multi-faceted in that it is able to address semiquantitatively the issue of disease burden, and also to assess disease physiology. For example, as firefly luciferase (F-Luc) activity is dependent on the presence of oxygen and ATP, photons are only emitted from metabolically active cells. Thus, therapeutic effects involving changes in tumor metabolism can be readily assessed, where necrotic regions within a tumor can be identified and potentially act as a marker for a positive drug response. The high sensitivity of BLI also allows for the detection of small numbers of tumor cells very early in the development of primary or metastatic disease; cancer cells can be visualized using BLI before they can be visualized by other imaging methods. Over a very short period of time, studies involving BLI have demonstrated that the technique is highly sensitive, high throughput, and relatively easy to use. It is likely that the use of BLI over the

In this chapter, the information gleaned from oncology research using BLI is summarized. The strengths, weaknesses and major findings from the last twenty years are consider in an attempt to exemplify the utility of BLI as well as illustrate some of its limitations. Major problems associated with this imaging modality are recognized in order to assist in designing preclinical experiments for those using this imaging modality. Topics summarized below include the development of luciferase positive orthotopic, metastatic and genetically engineered models of human cancer as well as the use of BLI for the assessment of therapeutic activity of drug candidates, as a tool for monitoring gene delivery and gene expression in vivo, for assessment of processes such as angiogenesis and apoptosis, and, finally, for imaging of metastasis and minimal disease in cancer models.

Bioluminescence (BL) is defined as the production of light by a living organism. Many organisms such as bacteria, fungi, fish, marine invertebrates, and insects use BL for the purpose of mating, camouflage, repulsion, communication and illumination. The chemical reaction that produces BL requires a pigment known generally as luciferin and enzymes called luciferase (see Reaction 1). The reaction between luciferase and its substrate is an oxidation reaction which is sometimes mediated by cofactors such as calcium and may

stages of tumor development in the same animal as the disease progresses.

next decade will continue to increase in its complexity and its elegance.

**2. Bioluminescence** 

require energy in the form of ATP.

of cancer cells. Thus if left untreated primary malignancies can evolve into a metastatic disease which ultimately engender systemic changes that are incompatible with life. Even this very simplistic description of cancer biology highlights the serious shortcoming of subcutaneous tumor models derived following injection of cultured tumor cell lines. The limitations are even more profound when considering changes in tumor biology that occur as a consequence of treatments.

In recent years orthotopic inoculation of tumor cells has be viewed as a reasonable alternative for initiation of model tumors. In these models, tumor cells are injected in a site that represents the tissue of origin and thus may be closer in characteristics to the original microenvironment. Orthotopic models can exhibit tumor growth rates, capacity for angiogenesis and an invasive potential that better mimic the evolution of cancers in situ. In several examples primary tumors arising following orthotopic injection of tumor cells have been shown to metastasize through lymphatic drainage and/or hemotological spread in ways that are comparable to that seen in human disease. For this reason, some investigators believe that orthotopic tumors more accurately reflect human disease, and may serve to better predict therapeutic outcomes.

Further, in an effort to model systemic disease a variety of cell inoculation methods can be employed to promote haematological spread of tumor cells. These methods include, but are not limited to, tail vein and intracardiac injections. Many groups have used different inoculation methods to assess specific metastatic sites, such as intratibial inoculations of prostate cancer cells in order to study bone metastases, or intraperitoneal injections of ovarian cancer cells to study disease development/progression in the peritoneal cavity. Finally, transgenic animal models are now readily available, where oncogenes and tumor suppressor genes relevant to a specific cancer are knocked in or knocked out. These genetically engineered animals have been useful for studies exploring how genetic alterations are linked to carcinogenesis.

A very large obstacle in studies using orthotopic, systemic and transgenic animal models is monitoring disease burden. The tumour tissue is often inaccessible for visual inspection, localized in organs deep within the body. Assessment of such tumors requires termination of animals at various time-points following disease initiation or at a time when the animal experiences signs of distress/illness. Mice are then euthanized and organs removed for gross and histological assessment of primary and metastatic disease; a practice that frequently requires serial sacrifice and large numbers of animals for a single study. Additionally, in this type of study design, comparisons are made between different groups of animals that were sacrificed at different time-points. Due to animal to animal variations, comparisons are often difficult to interpret and conclusions may be over- or even understated.

The use of orthotopic and transgenic cancer models has fostered development of smallanimal imaging methods to follow tumour development and progression in live animals. There are several modalities that are applicable to small animal imaging, including ultrasound (US), magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), and single photon emission computed tomography (SPECT). Each of these imaging modalities have strengths and weaknesses as recently reviewed by Ray et al (Ray 2011). Over the last decade use of Bioluminescence Imaging (BLI) has become

of cancer cells. Thus if left untreated primary malignancies can evolve into a metastatic disease which ultimately engender systemic changes that are incompatible with life. Even this very simplistic description of cancer biology highlights the serious shortcoming of subcutaneous tumor models derived following injection of cultured tumor cell lines. The limitations are even more profound when considering changes in tumor biology that occur

In recent years orthotopic inoculation of tumor cells has be viewed as a reasonable alternative for initiation of model tumors. In these models, tumor cells are injected in a site that represents the tissue of origin and thus may be closer in characteristics to the original microenvironment. Orthotopic models can exhibit tumor growth rates, capacity for angiogenesis and an invasive potential that better mimic the evolution of cancers in situ. In several examples primary tumors arising following orthotopic injection of tumor cells have been shown to metastasize through lymphatic drainage and/or hemotological spread in ways that are comparable to that seen in human disease. For this reason, some investigators believe that orthotopic tumors more accurately reflect human disease, and may serve to

Further, in an effort to model systemic disease a variety of cell inoculation methods can be employed to promote haematological spread of tumor cells. These methods include, but are not limited to, tail vein and intracardiac injections. Many groups have used different inoculation methods to assess specific metastatic sites, such as intratibial inoculations of prostate cancer cells in order to study bone metastases, or intraperitoneal injections of ovarian cancer cells to study disease development/progression in the peritoneal cavity. Finally, transgenic animal models are now readily available, where oncogenes and tumor suppressor genes relevant to a specific cancer are knocked in or knocked out. These genetically engineered animals have been useful for studies exploring how genetic

A very large obstacle in studies using orthotopic, systemic and transgenic animal models is monitoring disease burden. The tumour tissue is often inaccessible for visual inspection, localized in organs deep within the body. Assessment of such tumors requires termination of animals at various time-points following disease initiation or at a time when the animal experiences signs of distress/illness. Mice are then euthanized and organs removed for gross and histological assessment of primary and metastatic disease; a practice that frequently requires serial sacrifice and large numbers of animals for a single study. Additionally, in this type of study design, comparisons are made between different groups of animals that were sacrificed at different time-points. Due to animal to animal variations, comparisons are often difficult to interpret and conclusions may be over- or even

The use of orthotopic and transgenic cancer models has fostered development of smallanimal imaging methods to follow tumour development and progression in live animals. There are several modalities that are applicable to small animal imaging, including ultrasound (US), magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), and single photon emission computed tomography (SPECT). Each of these imaging modalities have strengths and weaknesses as recently reviewed by Ray et al (Ray 2011). Over the last decade use of Bioluminescence Imaging (BLI) has become

as a consequence of treatments.

better predict therapeutic outcomes.

alterations are linked to carcinogenesis.

understated.

increasingly popular in part because of the accessibility of imaging tools and because the method is extremely sensitive with a capability of detecting as few as 10 tumor cells in a live animal. BLI provides a non-invasive, semi-quantitative approach to localizing small tumors, following growth and metastasis and monitoring tumor response to treatment in the same animal longitudinally. This non-invasive determination of tumor burden over time reduces the numbers of animals required for experiments and provides information on the various stages of tumor development in the same animal as the disease progresses.

The first experiments using BLI for monitoring tumor phenotypes and response to therapy were performed by assessing luciferase activity as a measure of metabolism as described in section 6.0. Recently, more complex studies have been designed. For example, our group used BLI to track the development of an experimental metastatic model of breast cancer after an intracardiac injection of tumor cells. Further, we evaluated the use of an existing and clinically relevant drug to treat orthotopic, metastatic and ascities disease and correlated changes in tumor burden as measured by BLI to pharmacokinetic data in the same animals. The utility of BLI in assessing drug efficacy is multi-faceted in that it is able to address semiquantitatively the issue of disease burden, and also to assess disease physiology. For example, as firefly luciferase (F-Luc) activity is dependent on the presence of oxygen and ATP, photons are only emitted from metabolically active cells. Thus, therapeutic effects involving changes in tumor metabolism can be readily assessed, where necrotic regions within a tumor can be identified and potentially act as a marker for a positive drug response. The high sensitivity of BLI also allows for the detection of small numbers of tumor cells very early in the development of primary or metastatic disease; cancer cells can be visualized using BLI before they can be visualized by other imaging methods. Over a very short period of time, studies involving BLI have demonstrated that the technique is highly sensitive, high throughput, and relatively easy to use. It is likely that the use of BLI over the next decade will continue to increase in its complexity and its elegance.

In this chapter, the information gleaned from oncology research using BLI is summarized. The strengths, weaknesses and major findings from the last twenty years are consider in an attempt to exemplify the utility of BLI as well as illustrate some of its limitations. Major problems associated with this imaging modality are recognized in order to assist in designing preclinical experiments for those using this imaging modality. Topics summarized below include the development of luciferase positive orthotopic, metastatic and genetically engineered models of human cancer as well as the use of BLI for the assessment of therapeutic activity of drug candidates, as a tool for monitoring gene delivery and gene expression in vivo, for assessment of processes such as angiogenesis and apoptosis, and, finally, for imaging of metastasis and minimal disease in cancer models.
