B. anthracis protective antigen: a molecular imaging agent targeting squamous cell carcinoma

Abstract

Targeted molecular imaging is an emerging non-invasive tool for the identification and characterisation of cancers based on physiological and molecular properties. Molecular imaging is likely to play a significant role in the development and validation of future biomarkers due to the ability to identify and monitor real time tumour characteristics non-invasively. There is currently a need to identify novel biomarkers that can be used to target, or inform on, tumour specific processes.The human anthrax toxin receptors Tumour Endothelial Marker 8 (TEM8) and Capillary Morphogenesis Gene 2 (CMG2) have been identified as potential novel tumour biomarkers. TEM8 and CMG2 are cell surface receptors involved in angiogenesis and extracellular matrix homeostasis, processes that are important for tumour growth and metastasis. TEM8 is particularly interesting due to its tumour specific upregulation and undetectable expression in many healthy adult tissues. One method to target these receptors is to exploit their exogenous ligand Protective Antigen (PA). PA is a protein produced by the bacterium B. anthracis that mediates binding and internalisation of toxins into cells. Whilst preclinical studies have shown that PA can be used to specifically target toxins to tumours, the tumour specific localisation of PA in vivo has yet to be demonstrated.We have expressed recombinant PA, using an E. coli expression system, and purified the protein by liquid chromatography techniques. PA was non-specifically labelled with an amine reactive near infra-red fluorescent dye for optical imaging or 89Zr-Desferoxamine (89Zr-DFO) for positron emission tomography (PET). The labelled PA was intravenously (IV) injected into female HPV38E6E7-FVB mice, a transgenic mouse model of squamous cell carcinoma (SCC) that endogenously produces tumours in response to repeated UV irradiation.The fluorescently labelled PA protein was highly specific to the SCC tumours, resulting in rapid tumour uptake and exclusive tumour identification 24 hours post injection. Post mortem analysis, 48 hours post injection, showed that the fluorescence in the tumour was significantly higher (p < 0.002, two tailed paired T test) than the adjacent skin, the liver, and the kidneys. Immunofluorescence analysis of formalin-fixed paraffin-embedded skin tissues identified for the first time that both TEM8 and CMG2 are upregulated in SCC tumours compared to healthy skin, with the administered PA protein more significantly colocalised with TEM8 expression than CMG2 expression. As TEM8 and CMG2 appear to be expressed within different structures of the skin, the greater localisation of PA to TEM8 expression may be due to improved accessibility from an i.v. administered agent.PET imaging of 89Zr radiolabelled PA also showed tumour uptake, reaching a maximum approximately 5 hours post injection. However, the tumour uptake was relatively low with the majority of activity present in the kidneys. Analysis of biodistribution revealed that PA was rapidly cleared from the blood through hepatic and renal clearance mechanisms, with significant radioactivity retained within the kidneys. Rapid clearance can be an advantageous attribute for a targeted imaging agent, allowing rapid assessment after administration. However, the non-specific uptake of radioactivity as the result of PA degradation is a limitation of the 89Zr-DFO labelling. Further development is needed to improve PA as a PET imaging agent, utilising other radionuclides or protein modifications.Due to its high specificity for receptors overexpressed on many tumours and its natural internalisation function, PA has the potential to be used for diagnostic and therapeutic targeting of cancers. Our results demonstrate upregulation of both anthrax toxin receptors on SCC tumours, with TEM8 exhibiting great accessibility from an i.v. administered PA imaging agent. PA has the potential to be developed as a systemic targeting agent of SCC. Further work is needed to assess the in vivo expression of the anthrax toxin receptors during tumour development, in order to characterise their role in tumourigenesis. The HPV38E6E7-FVB mouse model is ideal for longitudinal studies biomarker assessment, developing tumours in an environment that is physiologically relevant and clinically translatable. Future work would benefit from characterization of how different labelling techniques and imaging molecules affect the biodistribution of PA in vivo. Additionally, a rationally designed molecule based on PA with improved biological properties could be developed for future cancer diagnostic and therapeutic applications.