In vivo bacterial imaging without engineering; A novel probe-based strategy facilitated by endogenous nitroreductase enzymes.

Abstract

The feasibility of utilising bacteria as vectors for gene therapy is becoming increasingly recognised. This is primarily due to a number of intrinsic properties of bacteria such as their tumour targeting capabilities, their ability to carry large genetic or protein loads and the availability of well-established genetic engineering tools for a range of common lab strains. However, a number of issues relating to the use of bacteria as vectors for gene therapy need to be addressed in order for the field to progress. Amongst these is the need for the development of non-invasive detection/imaging systems for bacteria within a living host. In vivo optical imaging has advanced preclinical research greatly, and typically involves engineering of bacteria with genetic expression constructs for luminescence (e.g. the lux operon) or fluorescent proteins (GFP etc.). This requirement for genetic modification can be restrictive, where engineering is not experimentally appropriate or technologically feasible (e.g. due to lack of suitable engineering tools). We describe a novel strategy exploiting endogenous bacterial enzymatic activity to specifically activate an exogenously administered fluorescent imaging probe. The red shifted, quenched fluorophore CytoCy5S is reduced to a fluorescent form by bacterial-specific nitroreductase (NTR) enzymes. NTR enzymes are present in a wide range of bacterial genera and absent in mammalian systems, permitting highly specific detection of Gram-negative and Gram-positive bacteria in vivo. In this study, dose-responsive bacterial-specific signals were observed in vitro from all genera examined - E. coli, Salmonella, Listeria, Bifidobacterium and Clostridium difficile. Examination of an NTR-knockout strain validated the enzyme specificity of the probe. In vivo whole-body imaging permitted specific, dose-responsive monitoring of bacteria over time in various infection models, and no toxicity to bacteria or host was observed. This study demonstrates the concept of exploiting innate NTR activity as a reporting strategy for wild-type bacteria using optical imaging, while the concept may also be extended to NTR-specific probes for use with other imaging modalities.